SMOS+SOS TS v2

SMOS+SOS Technical Specification Version 2.0

Code SMOS+SOS_TS Version 2.0 Date 24/06/14

Client European Space Agency Contract 4000107208/12/I-AM

ESA Support to Science Element (STSE)

SMOS+ Surface Ocean Salinity and Synergy

Technical Specification (TS) D-90

Prepared by LOCEAN-IFREMER/CATDS research team (France)

in partnership with

National Oceanography Centre (UK), Met Office (UK) and

Satellite Oceanographic Consultants Ltd (UK)

Name Signature Date

Author Jacqueline Boutin

24 June 2014

Approved Ellis Ash 24 June 2014

Revised

Authorised Craig Donlon

24 June 2014 i Technical Specification

Contents 1 Introduction ...............................................................................................................................1

2 Data Access and Requirements............................................................................................2 2.2 Data description tables .....................................................................................................................4 2.2.1 Satellite Data.....................................................................................................................................................4 2.2.2 In-‐situ Data..................................................................................................................................................... 24 2.1.1 Model Output ............................................................................................................................................ 41

2.2 Data Design Justification ........................................................................................................... 44 2.3 Risks and proposed solutions.................................................................................................. 44

3 Algorithm Theoretical Basis Description...................................................................... 45 Case study 1: Amazon/Orinoco plumes........................................................................................... 45 Case study 2: Agulhas, Gulf Stream ................................................................................................... 48 Case study 3: Tropical Pacific & Atlantic......................................................................................... 49 Case study 4: Sub-tropical North Atlantic (SPURS)...................................................................... 53 Case study 5: Equatorial Pacific ......................................................................................................... 58

4 Tools and systems to be implemented........................................................................... 60 4.1 Tools and systems by partner.................................................................................................. 60 4.1.1 LOCEAN systems ..................................................................................................................................... 60 4.1.2 Ifremer systems ....................................................................................................................................... 61 4.1.3 NOC systems.............................................................................................................................................. 61 4.1.4 MetOffice systems................................................................................................................................... 61

4.2 Tools Design Justification.......................................................................................................... 62 4.3 Risks and proposed solutions.................................................................................................. 62

5 System development requirements................................................................................ 63 5.1 Development requirements by Case Study Experiment ................................................. 63 5.2 Development Design Justification .......................................................................................... 64 5.3 Risks and proposed solutions.................................................................................................. 64

6 Interface control for the project....................................................................................... 65 6.1 Data interface matrix.................................................................................................................. 65 6.2 Experimental Scientific Dataset .............................................................................................. 68 6.2.1 CS1 Scientific Dataset ............................................................................................................................ 68 6.2.2 CS2 Scientific Dataset ............................................................................................................................ 68 6.2.3 CS3 Scientific Dataset ............................................................................................................................ 69 6.2.4 CS4 Scientific Dataset ............................................................................................................................ 69 6.2.5 CS5 Scientific Dataset ............................................................................................................................ 70

6.3 Interface Design Justification................................................................................................... 70 6.4 Risks and proposed solutions.................................................................................................. 70

7 Risk analysis and mitigation.............................................................................................. 71

8 Summary and conclusions.................................................................................................. 72

9 References ............................................................................................................................... 73

10 Appendices............................................................................................................................ 78

24 June 2014 ii Technical Specification

Change history Issue Date Comments Author

1.1 10/07/13 Initial release Jacqueline Boutin + Team

2.0 24/06/14 Addition of information on data and methods in section 3

Jacqueline Boutin + Team

Acronyms and Abreviations Aquarius Salinity mission of NASA/CONAE

CATDS Centre Aval de Traitement des Données SMOS

CLIVAR

CONAE

Climate Variability and Predictability

Comisión Nacional de Actividades Espaciales (Space Agency of Argentina)

CS Case Study

DS Data Set

EDS Experimental Data Set

ESA European Space Agency

FOAM(/NEMO) UK Met Office Forecasting Ocean Assimilation Model using NEMO

GCOS Global Climate Observing System

IPCC Intergovernmental Panel on Climate Change

ITCZ Intertropical Convergence Zone

NASA (US) National Aeronautics and Space Administration

NCOF (UK) National Centre for Ocean Forecasting

NEMO Nucleus for European Modelling of the Ocean

NERC (UK) Natural Environment Reseach Council

NOC National Oceanography Centre

OS Ocean Salinity

OSD Ocean Salinity at depth (deeper than 10 m)

OSMOSIX Oceanography using SMOS for innovative air-sea eXchanges studies (SMOS+SOS)

OSn Ocean salinity at depth n metres

PM Progress Meeting

PVR Project Validation Report

SAP Scientific Analysis Plan

SatOC Satellite Oeanographic Consultants

SCT Special Conditions of Tender

SMOS Soil Moisture and Ocean Salinity mission

SMOS-MODE SMOS Mission Oceanographic Data Explotation (ESF COST Action)

SOS Surface Ocean Salinity and Synergy (project)

24 June 2014 iii Technical Specification

SoW Statement of Work

SSS Sea surface salinity

SST Sea surface temperature

STSE Support to Science Element

TS Technical Specifications

UKMO UK Met Office

UM User Manual

UOS Upper ocean salinity (in top 10 metres)

WP Work Package

WWW World Wide Web

24 June 2014 1 Technical Specification

1 Introduction

This document provides the Technical Specification (TS) of ESA STSE SMOS+SOS project. It provides a detailed description of what and how the systems/algorithms/processing chains will be developed and implemented in Task 4 (Case Study Data Set Collection) and Taks 5 (Case Study System Implementation, Product Development and Validation) to be used in Task 6 (Case Study Scientific Analysis) of the project. It provides an overview of the end-to-end technical solution that will be implemented to address the technical requirements of the SAP and successfully perform all SSS case study experiments. It describes the overall data flow within the project and identifies major risks and proposes solutions.

This document should be read in conjunction with the Scientific Analysis Plan (SAP) which captures a detailed description of the scientific analysis to be performed as five Case Studies (CS) throughout the project, as summarised below.

Ref. Case Study area Motivation Approximate Co-ordinates

CS1 Amazon/Orinoco plumes Freshwater Outflow 70°W-30°W 5°S-30°N

CS2 Agulhas, Gulf Stream Strong water mass boundary region 75º–40ºW/5ºE–50ºE 30ºN–50ºN/25ºS–40ºS

CS3 Tropical Pacific & Atlantic Strong precipitation regime 140°E-90°W 30°S-30°N

CS4 Sub-tropical North Atlantic (SPURS) Strong evaporative regime 30ºW-45ºW 15ºN-30ºN

CS5 Equatorial Pacific Equatorial upwelling 140°E-90°W 5°S-5°N


2 Data Access and Requirements

This section is intended to document all satellite, in situ, model output and other data sets required by the project following the template provided in Appendix-1 of the SoW. Each data source has been given a unique Identifier, the data sources itemised, a simple description of the data set is given and the relationship to the SAP is traced. Detailed descriptions of each data et are provided in Section 2.2 of the report as required by the SoW.

2.1 Summary of Data Sources

Label Data source Description Used in Case Study Experiments :

Used for SAP requirements :

Satellite (EO) data sources

EO-01 SMOS L3 : CATDS CEC

SMOS L3 CEC product developed at CATDS

CS1-10, CS1-20, CS1-30, CS2-10, CS2-20, CS2-30, CS4-30, CS5-10, CS5-20, CS5-30

1-10, 2-25, 4-65

EO-02

SMOS L3 CATDS-CEC -LOCEAN

SMOS L3 products based on ESA L2 v5 products

CS3-10, CS4-30, CS5-10, CS5-20, CS5-30

4-10, 4-60

EO-03 SMOS L2 SMOS L2 product from ESA v5

CS2-10, CS2-20, CS2-30, CS3-10,CS3-20,CS3-30,CS4-10, CS4-30,CS5-20, CS5-30

2-30,

EO-04 AQUARIUS-NASA

Aquarius L2 & L3 products provided by NASA

CS1-10, CS4-30 1-15, 4-25

EO-05 AQUARIUS L3-NOC

Aquarius L3 products built by NOC from Aquarius L2 NASA products

CS2-10, CS2-20, CS2-30, CS4-20

2-40, 4-40

EO-06 AMSR-E AMSRE L2 used to derive AMSRE SSS

CS1-10, CS1-30 1-85

EO-07 MODIS Ocean colour from MODIS/Aqua satellite (used in case Globcolor fields not available)

CS1-40, 1-30 1-85, 1-90

EO-08 GLOBCOLOR Ocean colour fields merged from various satellites

CS1-40 1-85, 1-90, 1-95


EO-09 OSCAR Currents

Surface currents derived from altimetry

CS1-30, CS2-20, CS2-30

1-50, 2-60

EO-10 SSM/I Rain Rain rate from SSM/I F16 & F17

CS3-10,CS3-20 3-10, 3-20, 3-40

EO-11 TRMM 3B42 3-hour Rain Rates products from various satellites

CS3-10,CS3-20, CS4-10

3-10, 3-20, 3-40, 4-20

EO-12 AVHRR (MetOp)

SST images for cloud-free days

CS2-20, CS2-30 2-20

EO-13 OSTIA SST CS2-20, CS2-30 2-15, 2-20

EO-14 ASCAT Wind speed CS1-30 1-60

In situ (IS) data sources

IS-01 ARGO S measured by standards profiling ARGO floats

CS1-10, CS1-20, CS2-10, CS3-10

1-20, 1-25, 1-35, 1-45, 2-45, 2-50, 3-25

IS-02 ARGO-STS S measured by high resolution profiling floats

CS3-20 3-15,3-30,3-35

IS-03 ORE-SSS Ship SSS data from the ORE-SSS database

CS2-10, CS4-30 2-70,4-80

IS-04 Drifters SSS measured between 50cm and the surface by surface drifters

CS3-20 3-30, 3-35

IS-05 ISAS Optimal interpolation maps of in situ SSS

CS3-10, CS4-30,CS5-10

4-70

IS-06 STRASSE TSG SSS onboard SPURS/STRASSE campaign

CS4-30 4-50

IS-07 TAO/TRITON Measurements made on TAO/TRITON moorings

CS5-20 5-30

IS-08 MAP CO2 pCO2 measured on TAO moorings

CS5-30 5-35, 5-40

IS-09 NODC SSS data base CS 1-20 1-40

IS-10 ORE HYBAM River discharges CS 1-30, CS 1-40 1-65

Numerical model (NM) data sources

NM-01 FOAM-Global S, T, SSH and velocity daily means and/or 3-hour values, ¼°

CS2-10,CS2-20,CS2-30, CS3-30, CS4-10, CS4-20

2-35, 2-50, 4-15

NM-02 FOAM-NAtl S, T, SSH and velocity daily means , 1/12°

CS2-10,CS2-20, CS4-10, CS4-20

2-35, 2-50, 4-15


2.2 Data description tables Data description tables cover each item given in section 2.1.

2.2.1 Satellite Data Concerning SMOS data , a pro and cons of various CATDS SMOS products is available onto http://www.catds.fr/content/download/68781/908673/file/OS_products_Differences_and_ProsCons.pdf. Below are detailed descriptions.

EO-01 SMOS L3 : CATDS CEC IFREMER Product Name SMOS L3 : CATDS CEC IFREMER v2

Data type SSS satellite data Source CATDS CEC Key Websites http://www.catds.fr Version Ifremer Dataset V02 Platform name SMOS

Platform characteristics Sensor(s) MIRAS Sensor type L-band radiometer Sensor key technical characteristics Interferometer Analysis characteristics From L1b Reconstructed on EASE grid

SSS retrieval: SSS (Tbx +Tby) Wind-Model: Model 2 Calibration: Single OTT + daily 5°x5° adjustment wrt SSS climato Flagging: interorbit consistency / RFI % Region of FOV considered: AFFOV only Average: Simple average after through filtering of inconsistent SSS

References to technical specifications documents

http://www.catds.fr/Products/Available-products-from-CEC-OS

Product format Netcdf – rectangular grid Data gridding 0.25°, 0.5°, 1° Data coverage: temporal Daily, 10-days, Monthly

Data coverage: spatial Global coverage Project Requirements Date required within project June 2010 – December 2012 Use within project June 2010 – December 2012 Reason for selection Subset or complete record needed Complete record

Data quality Data calibration 1 OTT + calibration wrt climatology

Data validation See CATDS news Product limitations Climatology calibration


Potential product upgrades -

Data availability

Available from June 2010

Availability time-scale To December 2012

Estimates of data quantity Product delivery Netcdf files from ftp site

Data reliability - space segment based on ESA L1b

Data reliability - ground segment CATDS CEC

Pricing The CATDS products are freely available on FTP upon request.

To get access, just send an email to [email protected] with

• your name • your institution/company/university • http://www.remss.com/your scope of interest : Soil

Moisture and/or Ocean Salinity

Access conditions

Formal agreements with data suppliers The CATDS data are freely distributed. However, when using these data in a publication, we request that the following acknowledgement be given : "These [daily/10 Days/monthly] Composite, [quarter/half/one] degree resolution SMOS SSS data were obtained from the Ocean Salinity Expertise Center (CECOS) of the CNES-IFREMER Centre Aval de Traitemenent des Donnees SMOS (CATDS), at IFREMER, Plouzane (France). 2010. V02, [list the dates of the data used]."

Third party redistribution.

Miscellaneous

Comments

EO-02 SMOS L3 CATDS-CEC-LOCEAN Product Name SMOS L3 CATDS-CEC-LOCEAN

Data type SSS satellite data Source CATDS Key Websites http://www.catds.fr Version Locean v2013

Platform name SMOS

Platform characteristics Sensor(s) MIRAS Sensor type L-band radiometer Sensor key technical characteristics Interferometer Analysis characteristics Built from ESA level 2

Flagging: L2OS ‘retrieval flags’ and L2OS RFI flag Average: Weighted by retrieval error and SSS equivalent resolution



http://www.catds.fr/Products/Available-products-from-CEC-OS

J. Boutin, N. Martin, G. Reverdin, X. Yin and F. Gaillard, Sea surface freshening inferred from SMOS and ARGO salinity: Impact of rain, Ocean Sci., 9, 183-192, doi:10.5194/os-9-183-2013, 2013.

X. Yin, J. Boutin, and P. Spurgeon, “First assessment of SMOS data over open ocean: part I Pacific Ocean,” IEEE Transactions on Geoscience and Remote Sensing, 10.1109/TGRS.2012.2188407, 2012.

Product format Netcdf – rectangular grid Data gridding 0.25° (100km x 100km averages) Data coverage: temporal Monthly, 10-days

Data coverage: spatial Global coverage Project Requirements Date required within project Jan 2010- Dec 2012 Use within project Jan 2010- Dec 2012 Reason for selection Subset or complete record needed

Data quality Data calibration

Data validation

Product limitations Potential product upgrades

Data availability

Available from Jan 2010

Availability time-scale

Estimates of data quantity Monthly data: around 3 Giga 10 Days data: around 10 Giga

Product delivery Netcdf files from ftp site

Data reliability - space segment Data reliability - ground segment

Pricing The CATDS products are freely available on FTP upon request.

To get access, just send an email to [email protected] with

• your name • your institution/company/university • your scope of interest : Soil Moisture and/or Ocean Salinity

Access conditions Formal agreements with data suppliers The CATDS data are freely distributed. However, when using these

data in a publication, we request that the following acknowledgement and a reference the above publications be given: "The LOCEAN_v2013 Sea Surface Salinity maps have been produced by LOCEAN/IPSL (UMR CNRS/UPMC/IRD/MNHN)


laboratory that participates to theOcean Salinity Expertise Center (CECOS) of Centre Aval de Traitement des Donnees SMOS (CATDS). This product is distributed by the Ocean Salinity Expertise Center (CECOS) of the CNES-IFREMER Centre Aval de Traitement des Donnees SMOS (CATDS), at IFREMER, Plouzane (France). "


Miscellaneous

Comments

EO-03 SMOS L2 Product Name SMOS L2

Data type SSS satellite data Source Key Websites https://earth.esa.int/web/guest/missions/esa-operational-

eo-missions/smos http://www.argans.co.uk/smos.

Version ESA v5 Platform name SMOS

Platform characteristics Sensor(s) Microwave Imaging Radiometer using Aperture Synthesis -

MIRAS Sensor type Passive microwave 2D-interferometer Sensor key technical characteristics L-band (21 cm-1.4 GHz) Analysis characteristics Tb: ESA L1c (reconstructed on ISEA-15km grid)

SSS retrieval: L2OS v5 (Dwell-line; iterative retrieval) Wind-Model: Model 1 Calibration: Variable OTT (~every 2 weeks) Region of FOV considered: EAFFOV provided 130Tb in AFFOV (~+/-300km from swath centre)


Product format Data gridding ISEA – 15 km grid Data coverage: temporal Daily Data coverage: spatial Global coverage Project Requirements Date required within project Use within project Reason for selection Subset or complete record needed

Data quality Data calibration See description of these data quality & limitations onto

https://earth.esa.int/c/document_library/get_file?&folderId=127856&name=DLFE-1514.pdf

Data validation

Product limitations


Potential product upgrades

Data availability

Available from Jan 2010


Estimates of data quantity Product delivery

Data reliability - space segment ESA

Data reliability - ground segment ESA

Pricing ESA delivery Access conditions

Formal agreements with data suppliers


Miscellaneous Comments

EO-04 AQUARIUS -NASA Product name Aquarius L2& Aquarius L3 released by NASA

Data type SSS satellite data

Source NASA

Key Websites http://podaac.jpl.nasa.gov/datasetlist?ids=Sensor&values=AQUARIUS_RADIOMETER

Version V2.0

Platform name Aquarius

Platform characteristics

Sensor(s)

Sensor type Passive microwave radiometer

Sensor key technical characteristics L-band (21 cm-1.4 GHz), Push-broom system with three feed-horns

Analysis characteristics


Product format HDF5

Data gridding elliptical footprints of 76x94 km; 84x121 km; and 97x157 km

Data coverage: temporal 2011–present

Data coverage: spatial Global coverage

Project Requirements Date required within project 1 September 2011 Use within project CS2-10, CS2-20, CS2-30, CS4-20, CS4-30 Reason for selection Baseline Aquarius product Subset or complete record needed Ascending/descending phase can be specified

Data quality Data calibration As provided by NASA

Data validation


Product limitations V2.0 likely to be superseded by revised versions during lifetime of SMOS+SOS project


Data availability Available from PO.DAAC

Availability time-scale Near real time

Estimates of data quantity

Product delivery ftp access requires registration with valid email address Data reliability - space segment

Data reliability - ground segment

Pricing Freely available

Access conditions Formal agreements with data suppliers None


Miscellaneous

Comments

EO-05 AQUARIUS L3-NOC Product name NOC Aquarius L3

Data type SSS gridded satellite data

Source Po.daac via NOC

Key Websites http://podaac.jpl.nasa.gov/dataset/AQUARIUS_L2_SSS

Version L2 v2.0

Platform name Aquarius


Sensor(s)


Sensor key technical characteristics L-band (21 cm-1.4 GHz), Push-broom system with three feed-horns



Product format Matlab MAT file

Data gridding Any regular grid (ie latitude and longitude step are equal) with minimum time-step of one day

Data coverage: temporal September 2011-present

Data coverage: spatial global

Project Requirements Date required within project September 2011 Use within project CS2-10, CS2-20, CS2-30, CS4-20, CS4-30 Reason for selection Subset or complete record needed Can be split my ascending/descending phase or combined.

Can be split by feed-horn or combined Data quality

Data calibration


Data validation

Product limitations Non-operational product produced an ad hoc basis


Data availability

Available from NOC Availability time-scale


Product delivery Request via email to [email protected]

Data reliability - space segment Data reliability - ground segment


Access conditions

Formal agreements with data suppliers Supplied as part of NOC contribution to SMOS+SOS Third party redistribution. Cannot be redistributed without consent of NOC

Miscellaneous

Comments

EO-06 AMSR-E Product name AMSR-E L2A

Data type 1

Resampled Brightness Temperatures

Source National Snow and Ice Data Center (NSIDC)

Key Websites http://nsidc.org/data/docs/daac/ae_l2a_tbs.gd.html

Version 3

Platform name AMSR-E


Sensor(s)


Sensor key technical characteristics forward-looking,

66 conically scanning radiometer operating at 55± incidence and 9 frequencies between 6.9

67 and 89 GHz.




Product format HDF-EOS

Data gridding

Data coverage: temporal Sep 2002-Oct 2011

Data coverage: spatial global

Project Requirements

Date required within project Jan 2010-0ct 2011

Use within project

Reason for selection SSS inversion for Large river plume

Subset or complete record needed Zoom on Tropical regions only

Data quality

Data calibration

Data validation

Product limitations Non-available for Nov 2011-present


Data availability

Available from NSIDC



Product delivery Already accessed

Data reliability - space segment



Access conditions


Product name AMSR-E L2B

Data type ocean swath product (Wentz and Meissner [2000])

=> contains SST, near-surface wind speed, columnar water vapor, columnar cloud liquid water, and quality flags.

Source National Snow and Ice Data Center (NSIDC)


Key Websites http://nsidc.org/data/docs/daac/ae_ocean_products.gd.html

Version 3

Platform name AMSR-E


Sensor(s)


Sensor key technical characteristics forward-looking,

66 conically scanning radiometer operating at 55± incidence and 9 frequencies between 6.9

67 and 89 GHz.



Product format HDF-EOS

Data gridding

Near-surface wind speed at 38 km and 21 km resolution

Columnar water vapor at 21 km resolution

Columnar cloud liquid water at 12 km resolution

Data coverage: temporal Sep 2002-Oct 2011

Data coverage: spatial


Date required within project Jan 2010-0ct 2011

Use within project CS1-20,30,40

Reason for selection SSS inversion for Large river plume

Subset or complete record needed Zoom on Tropical regions only

Data quality

Data calibration

Data validation

Product limitations Non-available for Nov 2011-present


Data availability


Available from NSIDC



Product delivery Already accessed



EO-07 MODIS Product name MODIS 9km Standard Products Data type Aerosol Optical Thickness 869 nm, 8-day

Angstrom Coefficient, 8-day

Chlorophyll a concentration

Colored Dissolved Organic Matter (CDOM) Index, 8-day

Diffuse attenuation coefficient at 490 nm 9km

Fluorescence Line Height (normalized), 8-day

Particulate Inorganic Carbon, 8-day

Particulate Organic Carbon, 8-day

Photosynthetically Available Radiation, 8-day

Source National Snow and Ice Data Center (NSIDC) Key Websites http://nsidc.org/data/docs/daac/ae_ocean_products.gd.html Version 3 Platform name MODIS-Aqua

Platform characteristics Sensor(s) Sensor type Moderate Resolution Imaging Spectroradiometer Sensor key technical characteristics Orbit: 705 km, 10:30 a.m. descending node (Terra) or 1:30 p.m. ascending node (Aqua),

sun-synchronous, near-polar, circular Scan Rate: 20.3 rpm, cross track Swath Dimensions: 2330 km (cross track) by 10 km (along track at nadir) Telescope: 17.78 cm diam. off-axis, afocal (collimated), with intermediate field stop Size: 1.0 x 1.6 x 1.0 m Weight: 228.7 kg Power: 162.5 W (single orbit average) Data Rate: 10.6 Mbps (peak daytime); 6.1 Mbps (orbital average)


Quantization: 12 bits Spatial Resolution: 250 m (bands 1-2)

500 m (bands 3-7) 1000 m (bands 8-36)

Design Life: 6 years Analysis characteristics References to technical specifications documents

Product format Netcdf Data gridding Data coverage: temporal 2002/07/04 - 2013/06/ Data coverage: spatial global Project Requirements Date required within project Jan 2010-present Use within project CS1-20,30,40 Reason for selection Investigate SSS/ocean color relationships in Large tropical

river plume Subset or complete record needed Zoom on Tropical regions only

Data quality Data calibration Data validation Product limitations Cloud masked Potential product upgrades Data availability Available from NSIDC Availability time-scale Estimates of data quantity Product delivery Already accessed Data reliability - space segment Data reliability - ground segment Pricing Freely available Access conditions Formal agreements with data suppliers Third party redistribution. Cannot be redistributed without consent of NASA/GSFC Miscellaneous Comments EO-08 GLOBCOLOR Product name GLOB_4KM Data type


Covers the merged MERIS/MODIS/SEAWIF Level-3 ocean color products in the time period 1997-today

Source Key Websites http://www.globcolour.info/ Version Platform name Envisat, Aqua, terra Platform characteristics Sensor(s) MERIS-MODIS-SEAWIFS Sensor type Imaging Spectrometer Instruments Sensor key technical characteristics Analysis characteristics Merged data sets References to technical specifications documents

Product format Netcdf Data gridding Data coverage: temporal 1997-now Data coverage: spatial global Project Requirements Date required within project Jan 2010-present Use within project CS1-20,30,40, CS2 Reason for selection Subset or complete record needed subsets

Data quality Data calibration Data validation Product limitations clouds Potential product upgrades

Data availability Available from ESA Availability time-scale Estimates of data quantity Product delivery Already accessed Data reliability - space segment Data reliability - ground segment Pricing Freely available Access conditions Formal agreements with data suppliers Third party redistribution. Cannot be redistributed without consent of ESA Miscellaneous Comments

EO-09 OSCAR Currents Product name Ocean Surface Current Analyses Realtime (OSCAR) Data type surface current products

Source NOAA Key Websites http://www.oscar.noaa.gov Version 3


Platform name Various altimeters,radiometers & scatterometers: ENVISAT/ENVISAT RA-2 METOP-A/AVHRR-3 NOAA-16/AVHRR-3 NOAA-17/AVHRR-3 NOAA-18/AVHRR-3 JASON-1/POSEIDON-2 JASON-1/JMR QUIKSCAT/SEAWINDS TOPEX/POSEIDON/TOPEX ALTIMETER TOPEX/POSEIDON/POSEIDON ALTIMETER TOPEX/POSEIDON/ TOPEX MICROWAVE RADIOMETER ERS-1/ERS-1 ALTIMETER ERS-2/ERS-2 Altimeter DMSP-F10/SSM/I DMSP-F13/SSM/I DMSP-F11/SSM/I DMSP-F14/SSM/I GFO/GFO Altimeter NOAA-14/AVHRR-2 NOAA-19/ AVHRR-3 OSTM/Jason-2/POSEIDON-3 OSTM/Jason-2/AMR GRACE/GRACE ACC GRACE/GRACE SCA GRACE/GRACE KBR

Platform characteristics Sensor(s) Sensor type Sensor key technical characteristics Analysis characteristics References to technical specifications documents

Product format Netcdf Data gridding Data coverage: temporal 1992-now Data coverage: spatial global Project Requirements Date required within project Jan 2010-present Use within project CS1-20,30,40, CS2 Reason for selection Good product and easily accessible! Subset or complete record needed subsets

Data quality Data calibration Data validation Product limitations Temporal resolution (every 5 days) Potential product upgrades

Data availability Available from NOAA Availability time-scale Estimates of data quantity


Product delivery Already accessed Data reliability - space segment Data reliability - ground segment Pricing Freely available Access conditions Formal agreements with data suppliers Third party redistribution. Cannot be redistributed without consent of G. Lagerloeff Miscellaneous Comments

EO-10 SSM/I Rain Product Name SSM/I Rain Rate

Data type Satellite Source NASA MEaSUREs Program. Key Websites http://www.remss.com Version Version 7 Platform name DMSP

Platform characteristics 16 and F17 orbits Sensor(s) SSM/I Sensor type Microwave radiometer Sensor key technical characteristics 19, 22, 37, 85 GHz passive radiometers Analysis characteristics This algorithm is a product of 20 years of refinements,

improvements, and verifications. While the algorithms have evolved over time, a substantial background to the radiative transfer function used to derive the geophysical parameters is described in the following papers


Wentz F. J. 1997, "A well-calibrated ocean algorithm for SSM/I", J. Geophys. Res., Vol. 102, No. C4, pg. 8703-8718.

Wentz, Frank J. and Roy W. Spencer, May 1, 1998, "SSM/I Rain Retrievals within a Unified All-Weather Ocean Algorithm", Journal of the Atmospheric Sciences, Vol. 55, pg. 1613-1627.

Wentz, Frank J. and Thomas Meissner, 2000, "AMSR Ocean Algorithm, Version 2", report number 121599A-1, Remote Sensing Systems, Santa Rosa, CA, 66 pp.

Product format Binary data files Data gridding Orbital data mapped to 0.25 degree grid Data coverage: temporal Daily, 3-day, weekly, monthly

Data coverage: spatial Global coverage Project Requirements Date required within project 2010 to 2014 Use within project CS3, CS4, CS5 Reason for selection Within -80mn, +40mn from SMOS passes Subset or complete record needed

Data quality Data calibration 0 to 250 = valid geophysical data

251 = missing wind speed due to rain, missing water vapor due to heavy rain


252 = sea ice 253 = observations exist, but are bad (not used in composite maps) 254 = no observations 255 = land mass

Data validation See papers

Product limitations


Data availability

Available from http://www.remss.com, ftp://ftp.ssmi.com/ssmi

Availability time-scale F16 SSMIS: Oct 2003 to present

F17 SSMIS : Dec 2006 to present

Estimates of data quantity 200M

Product delivery download by ftp

Data reliability - space segment not known

Data reliability - ground segment not known

Pricing Free

Access conditions

Formal agreements with data suppliers Permission is granted to use these data and images in research and publications when accompanied by the appropriate instrument or product specific statement. : SSM/I data are produced by Remote Sensing Systems and sponsored by the NASA Earth Science MEaSUREs DISCOVER Project. Data are available at www.remss.com.

Third party redistribution. e.g. not permitted

Miscellaneous

Comments

EO-11 TRMM 3B42 TRMM and Other Satellites’ (3B42) products data

Tropical Rainfall Measuring Mission (TRMM).

Data type PRECIPITATION rate

Source NASA/Giovanni server

Key Websites http://disc2.nascom.nasa.gov/Giovanni/tovas/

http://disc.sci.gsfc.nasa.gov/precipitation/documentation/TRMM_README/TRMM_3B42_readme.shtml

Version 7 Platform name Tropical Rainfall Measuring Mission

Platform characteristics The Tropical Rainfall Measuring Mission (TRMM) is a joint U.S.-Japan satellite mission to monitor tropical and subtropical precipitation and to estimate its associated latent heating. TRMM was successfully launched on November 27, at 4:27 PM (EST) from the Tanegashima Space Center in Japan

Sensor(s) TRMM Precipitation Radar, TRMM Microwave Imager, TRMM


Visible Infrared Scanner

Sensor type Sensor key technical characteristics Analysis characteristics The purpose of the 3B42 algorithm is to produce TRMM-adjusted

merged-infrared (IR) precipitation and root-mean-square (RMS) precipitation-error estimates. The algorithm consists of two separate steps. The first step uses the TRMM VIRS and TMI orbit data (TRMM products 1B01 and 2A12) and the monthly TMI/TRMM Combined Instrument (TCI) calibration parameters (from TRMM product 3B31) to produce monthly IR calibration parameters. The second step uses these derived monthly IR calibration parameters to adjust the merged-IR precipitation data, which consists of GMS, GOES-E, GOES-W, Meteosat-7, Meteosat-5, and NOAA-12 data.


http://disc.sci.gsfc.nasa.gov/precipitation/documentation/TRMM_README/TRMM_3B42_readme.shtml

Product format Data gridding Horizontal Resolution: 0.25 degree x 0.25 degree

Vertical Resolution: Surface

Data coverage: temporal 3-hourly temporal or daily (derived from 3-hourly product)

Data coverage: spatial global belt (−180°W to 180° E) extending from 50°S to 50°N

latitude.

Project Requirements Date required within project 2010 to 2014 Use within project Reason for selection To study the link between SSS and rain rate Subset or complete record needed ?

Data quality Data calibration

Data validation http://trmm-fc.gsfc.nasa.gov/trmm_gv/information/brochure/brochure.html

Product limitations


Data availability

Available from http://disc2.nascom.nasa.gov/Giovanni/tovas/ http://mirador.gsfc.nasa.gov/cgi-bin/mirador/presentNavigation.pl?tree=project&project=TRMM

Availability time-scale 1998 to present


Product delivery ftp download

Data reliability - space segment not known

Data reliability - ground segment not known

Pricing Free

Access conditions


The *data access policy* is "freely available" with three common-sense caveats: 1. The data set source should be acknowledged when the data are used. The International Polar Year (IPY) Data policy guidelines (http://.ipydis.org/data/citations.html) suggest


a formal reference of the form Huffman, G.J., E.F. Stocker, D.T. Bolvin, E.J. Nelkin, R.F. Adler, 2012, last updated 2012: TRMM Version 7 3B42 and 3B43 Data Sets. NASA/GSFC, Greenbelt, MD. As an “Acknowledgment”, one possible wording is: "The TMPA data were provided by the NASA/Goddard Space Flight Center's Mesoscale Atmospheric Processes Laboratory and PPS, which develop and compute the TMPA as a contribution to TRMM." 2. New users should obtain their own current, clean copy, rather than taking a version from a third party that might be damaged or out of date. 3. Errors and difficulties in the dataset should be reported to the dataset creators.


Miscellaneous

Comments

EO-12 AVHRR (Metop) Product Name MetOp AVHRR No. Data type Level 2 sea surface temperature data Source EUMETSA OSI-SAF. Key Websites http://www.osi-saf.org/ Version 2.3 Platform name MetOp-A

Platform characteristics MetOp is a Low Earth Orbiting satellite operated by EUMETSAT. Sensor(s) AVHRR Sensor type Sensor key technical characteristics - Analysis characteristics - References to technical specifications documents

Low Earth Orbiter Sea Surface Temperature Product User Manual, vn 2.3, (http://www.osi-saf.org/production/cms/presentation_sst_leo.php?safosi_session_id=45bc8e4f44ad5776b8257ce8b368dee2#0)

Product format Netcdf GHRSST Data Specification v 2.0. https://www.ghrsst.org/documents/q/category/gds-documents/operational/

Data gridding The data is available at full resolution (~1km) MetOp SST in satellite projection. It is sub-sampled before storage at the UK Met Office using every 6th point in the across- and along-swath directions.

Data coverage: temporal Near real time operational data is available at the Met Office from 2007.

Data coverage: spatial Global Project Requirements Date required within project 2010-2012 . Use within project For comparison of Gulf Stream and Agulhas gradient features.

Reason for selection Data complements the OSTIA analysis as it can be used under clear sky conditions to view the gradients in the “raw” data without the smoothing induced by the OSTIA analysis procedure, but OSTIA is useful as it is available under all conditions and as a gridded field.


Subset or complete record needed Gulf Stream region; Agulhas region.

Data calibration Le Borgne, P., G. Legendre and A. Marsouin, (2007) Operational

SST retrieval from METOP/AVHRR, proceedings of the 2007 EUMETSAT conference, Amsterdam, The Netherlands, September 2007.

Data validation http://www.osi-saf.org/production/cms/validation_sst_leo.php?safosi_session_id=45bc8e4f44ad5776b8257ce8b368dee2

Product limitations -


Data availability

Available from The data is available over EUMETCAST. For the purposes of this project, the data will be used from internally at the Met Office.

Availability time-scale 2007 to now

Estimates of data quantity ~150MB per day.

Product delivery Internal to Met Office.

Data reliability - space segment See http://www.osi-saf.org/

Data reliability - ground segment See http://www.osi-saf.org/

Pricing The data are available free of charge through EUMETSAT.

Access conditions The data can be obtained without charge through the MyOcean service desk after signing an SLA.

Formal agreements with data suppliers -

Third party redistribution. -

Miscellaneous

Comments

EO-13 OSTIA SST Product Name OSTIA No. Data type Level 4 sea surface temperature and sea-ice objective analysis Source Various GHRSST L2p data-sets are used to produce this analysis

at the UK Met Office. Key Websites http://ghrsst-pp.metoffice.com/pages/latest_analysis/ Version 2.0 Platform name NOAA-19, MetOpA, MSG2, TRMM, GOES13, DMSP-F17,

DMSP-F15, in situ

Platform characteristics A mixture of polar orbiting and geostationary data, plus in situ surface temperature data.

Sensor(s) AVHRR, AVHRR_GAC, AVHRR_LAC, IASI, SEVIRI, TMI, GOES_Imager, SSMIS, SSM/I, ships, surface drifters, moored buoys,

Sensor type A mixture of infrared and microwave satellite instruments, and in situ sensors.

Sensor key technical characteristics - Analysis characteristics OSTIA uses satellite data provided by the GHRSST project,

together with in-situ observations to determine the sea surface temperature. The analysis is produced daily at a resolution of 1/20° (approx. 5km). OSTIA data is provided in GHRSST netCDF


format every day. References to technical specifications documents

Donlon, C.J., M. Martin, J. D. Stark, J. Roberts-Jones, and E. Fiedler, 2012. The Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA) system. Rem. Sens. of Environment. 116: 140-158. http://dx.doi.org/10.1016/j.rse.2010.10.017

Product format Netcdf GHRSST Data Specification v 2.0. https://www.ghrsst.org/documents/q/category/gds-documents/operational/

Data gridding The data is on 1/20° latitude/longitude grid. This is done using a multi-scale optimal interpolation-type technique described in Donlon et al 2012.

Data coverage: temporal Near real time operational data is available from 2006 – present. Reanalysis v1.0 is available from 1985-2007. Reanlaysis data produced as part of the ESA CCI project is available from 1991 – 2010. This data-set includes different data sources to those listed above (input data-sets produced as part of the CCI project, without any in situ data).

Data coverage: spatial Global Project Requirements Date required within project 2010-2012 . Use within project For comparison of Gulf Stream and Agulhas gradient features.

Reason for selection Complete coverage. More recent data has good feature resolution and its characteristics are robust to data sparseness (see Reynolds et al., 2013).

Subset or complete record needed Gulf Stream region; Agulhas region.

Data calibration See:

• Donlon, C.J., M. Martin, J. D. Stark, J. Roberts-Jones, and E. Fiedler, 2012. The Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA) system. Rem. Sens. of Environment. 116: 140-158. http://dx.doi.org/10.1016/j.rse.2010.10.017

• Matthew Martin, Prasanjit Dash, Alexander Ignatov, Viva Banzon,Helen Beggs, Bruce Brasnett, Jean-Francois Cayula, James Cummings, Craig Donlon, Chelle Gentemann, Robert Grumbine, Shiro Ishizaki, Eileen Maturi, Richard W. Reynolds and Jonah Roberts-Jones, Group for high resolution sea surface temperature (GHRSST) analysis fields inter-comparisons: Part 1. A GHRSST multi-product ensemble (GMPE), Deep-Sea Research II, http://dx.doi.org/10.1016/j.dsr2.2012.04.013

• Dash, P., et al., Group for High Resolution Sea Surface Temperature (GHRSST) analysis fields inter-comparisons -- Part 2: Near real time web-based based level 4 SST Quality Monitor (L4-SQUAM), Deep- Sea Res. II (2012), http://dx.doi.org/10.1016/j.dsr2.2012.04.002

Data validation http://ghrsst-pp.metoffice.com/pages/latest_analysis/sst_monitor/index.html



Data availability

Available from UK Met Office (for use internally), or through MyOcean service desk ([email protected]), or through



Estimates of data quantity 15MB per daily file.

Product delivery Internal to Met Office or via ftp.

Data reliability - space segment See Donlon et al., 2012.

Data reliability - ground segment See Donlon et al., 2012.

Pricing The data can be obtained without charge through the MyOcean service desk after signing an SLA.

Access conditions The data can be obtained without charge through the MyOcean service desk after signing an SLA.



Miscellaneous

Comments

EO-14 ASCAT wind speeds Product name Daily ASCAT Surface Wind Fields.

Data type 10-m height Wind speed components Source Ifremer/Cesrat Key Websites http://cersat.ifremer.fr/News/Products-informations/New-

Metop-ASCAT-wind-fields Version Platform name Metop/ASCAT

Platform characteristics Sensor(s) Sensor type Sensor key technical characteristics

Analysis characteristics These daily wind fields from Metop/ASCAT scatterometer retrievals are produced in near real-time over global ocean with a spatial resolution of 0.25°. They are in netCDF format and span from April 2007 to present/ New gridded daily wind and wind stress fields have been estimated over global oceans from Metop/ASCAT retrievals using objective method. The analyses use standard products ASCAT L2b during the period April 2007 through March 2009, and ASCAT L2b 12.5 from April 2009 to present. The resulting fields have spatial resolutions of 0.25° in longitude and latitude. The calculation of daily estimates uses ascending as well as descending available and valid retrievals. The objective method aims to provide daily-averaged gridded wind speed, zonal component, meridional component, wind stress and the corresponding components at global scale. The error associated to each parameter, related to the sampling impact and wind space and time variability, is provided too. More details about data, objective method, computation algorithm may be found in (Bentamy et al, 2011). The daily wind fields are calculated in near real time with a


delay of 48 hours.


ftp://ftp.ifremer.fr/ifremer/cersat/products/gridded/MWF/L3/ASCAT/Daily/

Product format Data gridding Horizontal Resolution: 0.25 degree x 0.25 degree

Vertical Resolution: Surface Data coverage: temporal daily

Data coverage: spatial global belt (−180°W to 180° E) extending from 50°S to 50°N

latitude. Project Requirements Date required within project 2007-now Use within project Reason for selection To study the link between SSS and air-sea exchanges Subset or complete record needed

subset

Data quality Data calibration Data validation ftp://ftp.ifremer.fr/ifremer/cersat/products/gridded/MWF/L3/A

SCAT/Daily/Doc/DailyAscatWind-Doc.pdf

Product limitations Potential product upgrades

Data availability Available from ftp://ftp.ifremer.fr/ifremer/cersat/products/gridded/MWF/L3/A

SCAT/Daily/Netcdf/

Availability time-scale 2007 to present Estimates of data quantity Product delivery ftp download Data reliability - space segment not known Data reliability - ground segment not known Pricing Free Access conditions Formal agreements with data suppliers

freely distributed

Third party redistribution. Miscellaneous Comments

2.2.2 In-situ Data

IS-01 ARGO LOCEAN/CATDS use ARGO data real time and when available delayed-time data delivered by CORIOLIS center (http://www.coriolis.eu.org/), metOffice use the EN3 data-set (http://www.metoffice.gov.uk/hadobs/en3/). Also some quality control check may slightly differ, this is not expected to have a big impact on our results, given the precision we need.


We describe below each product; for simplicity both will be called ARGO in the rest of the documents. Product Name Argo-CORIOLIS data set No. Data type 3,000 free-drifting temperature/salinity profiling floats Source Key Websites http://www.coriolis.eu.org/

Version Platform name -

Platform characteristics see http://www.argo.ucsd.edu/How_Argo_floats.html

Sensor(s)

Sensor type Depend on ARGO float type; see http://www.argo.ucsd.edu/How_Argo_floats.html

Sensor key technical characteristics Depend on ARGO float type; see http://www.argo.ucsd.edu/How_Argo_floats.html

Analysis characteristics See http://www.coriolis.eu.org/Data-Services-Products/Documentation


See http://www.coriolis.eu.org/Data-Services-Products/Documentation

Product format Netcdf Data gridding Data coverage: temporal Data coverage: spatial Global - around 3000 Project Requirements Date required within project 2010-2011 Use within project Ground truth for SMOS data Reason for selection Excellent sampling of SSS gradients Subset or complete record needed complete

Data calibration http://www.pmel.noaa.gov/tao/proj_over/sampling.htmlhtt

p://www.argodatamgt.org/Documentation/Validation-and-reference-data

Data validation http://www.argodatamgt.org/Documentation/Validation-and-reference-data&http://www.coriolis.eu.org/Data-Services-Products/Documentation Temperatures in the Argo profiles are accurate to ± 0.005°C and depths are accurate to ± 5m. For real time salinity, ~ ± .01 psu. It can be improve in some cases.

Product limitations Argo near real-time data is subject to only coarse fully-automated quality control checks. No quality control is done on Oxygen measurements at present. Argo delayed-mode procedures for checking sensor drifts and offsets in salinity rely on a statistical comparison of the float data


with reference data. An adjustment is made when the float PI judges that it will improve the quality of the dataset. Users should include the supplied error estimates in their usage of Argo delayed-mode salinity data.


Data availability

Available from Coriolis

Availability time-scale From 2000


Product delivery Coriolis : ftp://ftp.ifremer.fr/ifremer/argo



Pricing Free

Access conditions http://www.coriolis.eu.org/

Formal agreements with data suppliers " These data were collected and made freely available by the International Argo Program and the national programs that contribute to it. (http://www.argo.ucsd.edu, http://argo.jcommops.org). The Argo Program is part of the Global Ocean Observing System." See http://www.coriolis.eu.org/for more info


Miscellaneous

Comments

Metoffice data At the Met Office we tend to use the EN3 data-set which contains consistently quality controlled sub-surface data including Argo. Product Name ARGO-Metoffice EN3 No. Data type Salinity and temperature data Source The sources of the data used to produce the most recent version of

the EN3 dataset were WOD05, GTSPP, Argo and the ASBO project.

Key Websites http://www.metoffice.gov.uk/hadobs/en3/ Version EN3_v2a Platform name -

Platform characteristics - Sensor(s) -

Sensor type - Sensor key technical characteristics - Analysis characteristics The profile files contain observed profile data for a particular

month and quality control decisions for those profiles. Some data thinning has been applied to the original profile data: a single profile for each day is averaged from the data from moored buoys that report very frequently and for data with high vertical resolution the profiles are subsampled to retain a maximum of 150


levels. Only one profile is stored within 0.2 degrees in latitude and longitude and 1 hour. Data from the ASBO project only have the day of observation so in this case only one profile is allowed per day. The profile that is retained is determined using a number of factors such as which has the most levels, which goes the deepest, the source catalogue (e.g. Argo delayed mode data are preferred to Argo realtime profiles). More information can be found in Ingleby and Huddleston (2007). The analysis files contain objective analyses produced using optimal interpolation of the profile data for a particular month and a persistence forecast background. More details can be found in Ingleby and Huddleston (2007). The analyses are produced on a regular 1 degree latitude/longitude grid and have 42 levels in the vertical. To reduce storage space not all latitudes are saved in the files.


Ingleby, B., and M. Huddleston (2007), Quality control of ocean temperature and salinity profiles—Historical and real‐time data, J. Mar. Syst., 65, 158–175. The altimetry quality control methodology is described in: Guinehut, S., C. Coatanoan, A.-L. Dhomps, P.-Y. Le Traon and G. Larnicol, 2009. On the use of satellite altimeter data in Argo quality control, Journal of Atmospheric and Oceanic Technology, 26, 395-402, DOI: 10.1175/2008JTECHO648.1

Product format Netcdf 2 types of data : profiles and objectives analyses

Data gridding For profiles files: Only one profile is stored within 0.2 degrees in latitude and longitude and 1 hour. For analysis files: 1 degree latitude/longitude grid and 42 vertical layers

Data coverage: temporal Data are available from 1950 to the present and there are separate files for each month.

Data coverage: spatial Global Project Requirements Date required within project 2010-2012 and only data at 5 m depth Use within project Ground truth for SMOS data

Reason for selection Excellent sampling of SSS gradients

Subset or complete record needed Tropical Pacific; SPURS region; Gulf Stream region

Data calibration See: Ingleby, B., and M. Huddleston (2007), Quality control of

ocean temperature and salinity profiles - Historical and real‐time data, J. Mar. Syst., 65, 158–175.

Data validation http://www.argodatamgt.org/Documentation/Validation-and-reference-dataSee : http://www.metoffice.gov.uk/hadobs/en3/en3_file_formats.html To make correct use of the profile data, the quality control flags must be read and applied (see website)



Data availability

Available from Met Office Hadley Centre observations datasets



Estimates of data quantity 1 Giga

Product delivery http://www.coriolis.eu.org/ftp://ftp.ifremer.fr/ifremer/argohttp://www.metoffice.gov.uk/hadobs/en3/data/download.html

Data reliability - space segment -

Data reliability - ground segment -

Pricing The data can be obtained without charge for private study or scientific research, but please do read the terms and conditions before use (http://www.metoffice.gov.uk/hadobs/en3/terms_and_conditions.html). Registering and leaving feedback will help us to improve the dataset in the future.

Access conditions Agreements

Formal agreements with data suppliers When publishing work using the data, please use the following citation: Ingleby, B., and M. Huddleston, 2007: Quality control of ocean temperature and salinity profiles - historical and real-time data. Journal of Marine Systems, 65, 158-175 10.1016/j.jmarsys.2005.11.019 The source should also be quoted in the acknowledgements section as www.metoffice.gov.uk/hadobs.


Miscellaneous

Comments

IS-02 ARGO-STS Product Name Argo-STS data set No. Data type ARGO profiling float equipped with a surface temperature and

salinity sensor (STS) Source Key Websites http://www.coriolis.eu.org/

Version Platform name ARGO STS Platform characteristics see http://www.seabird.com/products/spec_sheets/41data.htm

http://www.argo.ucsd.edu/How_Argo_floats.html

Sensor(s)

Sensor type Seabird STS Sensor key technical characteristics http://www.seabird.com/technical_references/SSalOceanSciences

Mar08-5Pages.pdf

Analysis characteristics See http://www.coriolis.eu.org/Data-Services-Products/Documentation


See http://www.coriolis.eu.org/Data-Services-Products/Documentation

Product format Netcdf Data gridding Data coverage: temporal


Data coverage: spatial Tropical regions. About 50 floats Project Requirements Date required within project 2010-2012 Use within project Ground truth for evaluating rain effect on salinity vertical

stratification and validate SMOS observations Reason for selection Excellent sampling of SSS gradients Subset or complete record needed complete

Data calibration The trade-off for achieving sensor stability over the lifespan of

the Argo floats was the forfeiture of very near surface temperature and salinity information to prevent ingestion of sea surface oils and contaminants into the conductivity cell. With increasing interest in measuring surface salinity in the world’s oceans, Sea-Bird developed an integrated system that continues to gather high accuracy salinity and temperature data using the pumped SBE 41CP CTD between 2000 – 5 decibars, while co-deploying a free-flow sensor called STS in the upper 20 db to capture the surface temperature and salinity conditions. STS is a free-flushing sensor, and is expected to ingest sea surface oils and contaminants that may alter the sensor calibration over time. To correct any drift in STS, both the SBE 41CP CTD and the STS take measurements near the float park depth and again in the upper ocean, just before the SBE 41CP pump is turned off, and then STS continues sampling through the ocean surface. STS data can then be corrected by recalibrating it based on the comparison measurements to the clean data from the SBE 41CP (see http://www.seabird.com/technical_references/AquariusPosterDec08_3Pages.pdf for more information)

Data validation http://www.coriolis.eu.org/Data-Services-Products/Documentation (some In-situ calibration results for the STS sensor indicate that the agreement of the primary CTD temperature and salinity measurements are 0.0003 ± 0.004 degrees C and -0.002 ± 0.005 psu)

Product limitations STS is a second, free-flushed, conductivity sensor, used in conjunction with the SBE 41CP CTD. Itscalibration is expected to drift from fouling as itmeasures up through the ocean surface film. Overlapping data acquired with both conductivitysensors allows the STS calibration to be adjusted,profile by profile, to the stable and accurate 41CP. Users should include the supplied error estimates in their usage of Argo delayed-mode salinity data.


Data availability

Available from Coriolis

Availability time-scale From 2009


Product delivery Coriolis : ftp://ftp.ifremer.fr/ifremer/argo




Pricing Free

Access conditions http://www.coriolis.eu.org/

Formal agreements with data suppliers " These data were collected and made freely available by the International Argo Program and the national programs that contribute to it. (http://www.argo.ucsd.edu, http://argo.jcommops.org). The Argo Program is part of the Global Ocean Observing System." See http://www.coriolis.eu.org/for more info


Miscellaneous

Comments These data are not yet in Coriolis data base but should be put before Fall 2013 ; if not we got an agreement from Steve Reiser to access the original files.

IS-03 ORE-SSS No. Observation service of SSS Data type Ship Source LEGOS Key Websites http://www.legos.obs-

mip.fr/observations/sss Version Delayed Mode

Platform name Ship names (e.g. Toucan; Colibri)

Platform characteristics Opportunity ships Sensor(s) TSG Sensor type Seabird Sensor key technical characteristics See Web site documentation Analysis characteristics Delayed mode data References to technical specifications documents

Henin, C. and J. Grelet, 1996. A merchant ship thermo-salinograph network in the Pacific Ocean. Deep-Sea Research I, 43 (11-12), 1833-1855. http://abstracts.congrex.com/scripts/jmevent/abstracts/FCXNL-09A02-1650845-1-Extended_Abst_Delcroix_etal_Oceanobs09_VF.pdf in Proceedings of the "Oceanobs'09: Sustained Ocean Observations and Information for Society" Conference, Venice, Italy, 21-25 September 2009, Hall, J., D.E. Harrison and D. Stammer, Eds., ESA Publication WPP-306, 2010. Alory et al., 2010. A global Voluntary Observing Ships Sea Surface Salinity network: data collection and quality control, in preparation.


Product format netcdf Data gridding no Data coverage: temporal Until 2011 (2012 validation ongoing) Data coverage: spatial Global (see web site) Project Requirements Date required within project 2010-2011 Use within project Ground truth for SMOS data Reason for selection Excellent sampling of SSS gradients Subset or complete record needed Tropical Pacific; SPURS region; Gulf

Stream region Data quality ~0.1 (depending on ships)

Data calibration See http://www.legos.obs-mip.fr/observations/sss/

Data validation See http://www.legos.obs-mip.fr/observations/sss/


Potential product upgrades 2012 data under validation.

Data availability Up to end 2011

Available from 2002

Availability time-scale about one year

Estimates of data quantity Few Megabytes depending on the campaign

Product delivery ftp

Data reliability - space segment -

Data reliability - ground segment -

Pricing Free

Access conditions acknowledge the ORE-SSS at LEGOS and copies should be sent to the Contacts: Gael ALORY [email protected] Or Thierry DELCROIX [email protected]



Miscellaneous

Comments -

IS-04 Drifters Product Name Surface drifters

No. Data type Salinity and temperature at 50 cm for Pacificgyre, Metocean and

ICM drifters; at 4 cm for Surpact drifter; at 15 cm for Surplas drifters

Source SVP drifters from Pacific Gyre, MetOcean and ICM laboratory + SVP drifters SURPLAS ans SURPACT built at LOCEAN (Paris)


Key Websites https://www.locean-ipsl.upmc.fr/~smos/drifters/

Version Platform name -

Platform characteristics - Sensor(s) For PG and ICM drifters: Unpumped SBE37 C/T unit and an

additional SST measurement along the float’s hull near 18 cm For Meteocean drifters: SBE47 C/T unit For Surplas or Surpact drifters: Valport C/T unit

Sensor type See below Sensor key technical characteristics Data uncertainty: around 0.01 PSU Analysis characteristics -


Reverdin, G., S. Morrisset, J. Boutin, and N. Martin, 2012. Rain-induced variability of near sea-surface T and S from drifter data. Journal of Geophysical Research 117, C2, http://dx.doi.org/10.1029/2011JC007549.

Reverdin, G., S. Morisset, D. Bourras, N. Martin, A. Lourenço, J. Boutin, C. Caudoux, J. Font, and J. Salvador. 2013. Surpact: A SMOS surface wave rider for air-sea interaction. Oceanography 26(1):48–57, http://dx.doi.org/10.5670/oceanog.2013.04.

Product format Ascii file Data gridding - Data coverage: temporal The SURPLAS and SURPACT drifter provides a value (average over 8”

for SURPLAS and average over 50'' for SURPACT) every 15 minutes of T (Temperature) and S (Salinity); the Pacificgyre SVP-BS drifter, a value every 30 minutes (average over 5 minutes), the Metocean SVP-BS drift-er, a value every hour (average of 7 values over 10 minutes), and the ICM/CISC provide values at the time of Argos transmissions (not averaged). Most of the drifters and floats transmit through Argos, although Metocean drifters since 2009 mostly transmit data (and a 3-hourly gps position) through iridium communication.

Data coverage: spatial http://www.locean-ipsl.upmc.fr/smois/driftershttp://www.locean-ipsl.upmc.fr/smois/driftersSee web site for drifter positions: www.locean-ipsl.upmc.fr/smos/drifters

Project Requirements Date required within project Drifters from January 2010 (beginning of SMOS) Use within project Ground truth for SMOS data Reason for selection Excellent sampling of SSS gradients Subset or complete record needed Tropical Pacific; SPURS region; Gulf Stream region

Data quality Data calibration After a thorough quality check, accuracy better than 0.1psu.

Data validation http://www.pmel.noaa.gov/tao/proj_over/qc.htmlS. Morisset, G. Reverdin, J. Boutin, N. Martin, X. Yin, F. Gaillard, P. Blouch, J. Rolland, J. Font, J. Salvador. Surface salinity drifters for SMOS validation. Mercator Ocean-Coriolis Quaterly Newsletter, April 2012.

Product limitations Data in real time not validated Data from died drifters validated with a precision between 0.1 and 0.01 PSU

Potential product upgrades Upgrade of data when drifters died or new drifters are deployed.

Data availability


Available from www.locean-ipsl.upmc.fr/smos/drifters

Availability time-scale Real time data

Estimates of data quantity 300 Mega

Product delivery Webpage: www.locean-ipsl.upmc.fr/smos/drifters Data



Pricing Free

Access conditions www.locean-ipsl.upmc.fr/smos/drifters

Formal agreements with data suppliers Cite locean web site and: Reverdin, G., S. Morrisset, J. Boutin, and N. Martin, Rain-induced variability of near sea-surface T and S from drifter data" Journal of Geophysical Research – Oceans, VOL. 117, C02032, doi:10.1029/2011JC007549, 2012.


Miscellaneous

Comments

IS-05 ISAS SSS ISAS MAPS In Situ Analysis System (ISAS) optimal interpolation Data type Global SSS maps are derived from delayed time quality checked in

situ measurements (ARGO and ship) by IFREMER/LPO, Laboratoire de physique des oceans

Source IFREMER/LPO, Laboratoire de physique des oceans F. Gaillard pers. com.

Key Websites http://wwz.ifremer.fr/lpo/SO-Argo-France/Products/Global-Ocean-T-S/Monthly-fields-2004-2010

Version D7CA2S0 re-analysis product Platform name Not relevant

Platform characteristics Sensor(s) Sensor type Sensor key technical characteristics Analysis characteristics Analysis products References to technical specifications documents

Method description in: http://wwz.ifremer.fr/lpo/SO-Argo-France/Products/Global-Ocean-T-S/Monthly-fields-2004-2010 Publications: Gaillard F, Autret E, Thierry V, Galaup P, Coatanoan C, Loubrieu T (2009) Quality 1575 control of large Argo datasets, J. Atmos. Ocean. Tech., 26, 337–351 Gaillard, F. and R. Charraudeau (2008): ISAS-V4.1b: Description of the method and user manual. Rapport LPO 08-03 Gaillard, F. (2010), ISAS Tool Version 5.3: Method and configuration. SO-ARGO report.

Product format NETCDF


Data gridding 0.25 °, 1 month

Data coverage: temporal 2004 to 2011 Data coverage: spatial Global coverage Project Requirements Date required within project 01/01/10 to 31/12/2011 Use within project CS 3 & 5 Reason for selection Ground truth for SMOS validation and evidence SMOS

contribution wrt existing in situ network Subset or complete record needed ?

Data quality Data calibration ?

Data validation ?

Product limitations Low resolution due to in situ data sampling Two scale lengths are considered in the optimal analysis: the first one is isotropic and equal to 300 km, the second one is set equal to 4 times the average Rossby radius of the area. As a result, we expect these maps being smoothed over about 3° in tropical areas

Potential product upgrades ?

Data availability

Available from data available on request to [email protected]


Estimates of data quantity 140Giga

Product delivery



Pricing Free

Access conditions Upon request : Contact Fabienne Gaillard ([email protected])



Miscellaneous

Comments

IS-06 STRASSE Product Name TSG data No. Data type Sea surface salinity (SSS) and sea surface temperature (SST) Source RV Thalassa Key Websites http://www.locean-ipsl.upmc.fr/smos/spurs/ Version - Platform name -

Platform characteristics - Sensor(s) SBE 21


Sensor type Sensor key technical characteristics Analysis characteristics - References to technical specifications documents

-

Product format Ascii file Data gridding Data coverage: temporal Each 5 min Data coverage: spatial SPURS region Project Requirements Date required within project All Strasse data Use within project Ground truth for SMOS data Reason for selection Excellent sampling of SSS gradients Subset or complete record needed SPURS region

Data quality Data calibration Calibration of SBE after the cruise, and salinity samples taken every

day

Data validation Validation of data still need to be done. Preliminary results indicates a correction of around 0.01 PSU (to be confirmed)

Product limitations


Data availability

Available from From August 161 to September 13 2012

Availability time-scale -

Estimates of data quantity 1M

Product delivery http://www.locean-ipsl.upmc.fr/smos/spurs/



Pricing Free

Access conditions



Miscellaneous

Comments

IS-07 TAO/TRITON TAO/TRITON array The Tropical Atmosphere Ocean (TAO) array (renamed the

TAO/TRITON array on 1 January 2000) consists of approximately 70 moorings in the Tropical Pacific Ocean, telemetering oceanographic and meteorological data to shore in real-time via the Argos satellite system

No. Data type -In situ mooring

-OS1m , OS5m , OS10m, precipitations

Source Support is provided primarily by the United States (National Oceanic and Atmospheric Administration) and Japan (Japan Agency for Marine-earth Science and TEChnology).


Key Websites http://www.pmel.noaa.gov/tao/data_deliv/frames/main.html http://www.pmel.noaa.gov/tao/proj_over/sampling.html

Version Platform name

Platform characteristics ATLAS (Autonomous Temperature Line Acquisition System) mooring

Sensor(s) See http://www.pmel.noaa.gov/tao/proj_over/sensors.shtml Sensor type See http://www.pmel.noaa.gov/tao/proj_over/sensors.shtml

Sensor key technical characteristics See http://www.pmel.noaa.gov/tao/proj_over/sensors.shtml ;

For OS : Sample rate: 1 per 10min Sample period: Instantaneous or 60min Data recorded in memory: 10 min or hourly Transmitted data: Daily mean or hourly mean

Analysis characteristics References to technical specifications documents

Gage, D. Halpern, M. Ji, P. Julian, G. Meyers, G.T. Mitchum, P.P. Niiler, J. Picaut, R.W. Reynolds, N. Smith, K. Takeuchi, 1998: The Tropical Ocean-Global Atmosphere (TOGA) observing system: A decade of progress. J. Geophys. Res., 103, 14, 169-14,240.

Bourles, B., R. Lumpkin, M.J. McPhaden, F. Hernandez, P. Nobre, E.Campos, L. Yu, S. Planton, A. Busalacchi, A.D. Moura, J. Servain, and J. Trotte, 2008: The PIRATA Program: History, Accomplishments, and Future Directions. Bull. Amer. Meteor. Soc., 89, 1111-1125.

McPhaden, M.J., G. Meyers, K. Ando, Y. Masumoto, V.S.N. Murty, M. Ravichandran, F. Syamsudin, J. Vialard, L. Yu, and W. Yu, 2009: RAMA: The Research Moored Array for African-Asian-Australian Monsoon Analysis and Prediction. Bull. Am. Meteorol. Soc., 90, 459-480, doi:10.1175/2008BAMS2608.1

http://www.pmel.noaa.gov/tao/proj_over/sensors.shtml

http://www.pmel.noaa.gov/tao/proj_over/sampling.html

Product format ASCII or NETCDF Data gridding - Data coverage: temporal 1987 to now Data coverage: spatial [120E : 80W, 15S:15N] Project Requirements Date required within project 01/01/10 to 31/12/2011 Use within project CS-3 and CS5 Reason for selection Ground truth for SMOS SSS Subset or complete record needed

Data quality Data calibration See

http://www.pmel.noaa.gov/tao/proj_over/sampling.html

Data validation http://www.pmel.noaa.gov/tao/proj_over/qc.html Test of data quality: If Q=1, data has been validated


If Q=2, data not validated, check when it is possible, if bias exist, ie, check if SSS at surface depth minus SSS at subsurface layer depth do not present a systematic bias (see Hénocq et al. 2010). If Q >=3: We do not use this data

Product limitations When Quality index Q is equal to 2 and only a surface layer is available, it is not possible to check if the data is good


Data availability

Available from NOAA, http://www.pmel.noaa.gov/tao/data_deliv/frames/main.html


Estimates of data quantity 10Mo

Product delivery Tar file from http://www.pmel.noaa.gov/tao/



Pricing Free

Access conditions Acknowledge the TAO Project Office of NOAA/PMEL if you use these data in publications. Also, they would appreciate receiving preprint and/or reprint of publications utilizing the data for inclusion in the TAO Project bibliography. Relevant publications should be sent to: TAO Project Office NOAA/Pacific Marine Environmental Laboratory 7600 Sand Point Way NE Seattle, WA 98115 Comments and questions should be directed to [email protected] .



Miscellaneous

Comments -

IS-08 MAP CO2 MAP CO2 moorings The open ocean moored CO2 network is still in its infancy, but is

slowly expanding into a global network of surface ocean and atmospheric CO2 observations that will make a substantial contribution to the production of seasonal CO2 flux maps for the global oceans. The long-term goal of this project is to populate the network of OCEAN Sustained Interdisciplinary Time-series Environment observation System (OceanSITES) so that CO2 fluxes will become a standard part of the global flux mooring network. This effort has been endorsed by the OceanSITES science team. The moored CO2 project contributes to NOAA's Program Plan For Building a Sustained Ocean Observing System for Climate by directly addressing key element (7) Ocean Carbon Network, but also provides a value added component to elements (3) Tropical Moored Buoys and (6) Ocean Reference Stations.

No. Data type -In situ mooring


-pCO2

Source Support is provided by the United States (National Oceanic and Atmospheric Administration)

Key Websites http://www.pmel.noaa.gov/co2/story/Open+Ocean+Moorings

Version NRT Platform name 0N, 110°W; 0N,125°W; 0°N, 140°W;0N, 155°W; 0N, 170°W;

0°N, 165°E; Platform characteristics MAP CO2 Sensor(s) See

http://www.pmel.noaa.gov/co2/story/Buoys+and+Autonomous+Systems

Sensor type See http://www.pmel.noaa.gov/co2/story/Buoys+and+Autonomous+Systems

Sensor key technical characteristics See http://www.pmel.noaa.gov/co2/story/Buoys+and+Autonomous+Systems;:

Analysis characteristics Every 3 hours for at least one year References to technical specifications documents

Product format ASCII or NETCDF Data gridding - Data coverage: temporal 2006 to now (verified until 2009) Data coverage: spatial [Eq 110°W; Eq 165°E] Project Requirements Date required within project 01/01/10 to 31/12/2012 Use within project CS5 Reason for selection Very good temporal and spatial sampling Subset or complete record needed 2010-2012

Data quality Near Real Time Data calibration Use of atmospheric pCO2 as reference

Data validation

Product limitations Unverified product (real time)


Data availability

Available from Real time data visible on NOAA, http://www.pmel.noaa.gov/co2/story/Open+Ocean+Moorings Verified data available on http://cdiac.ornl.gov/oceans/Moorings/Pacific.html

Availability time-scale See http://www.pmel.noaa.gov/co2/story/Open+Ocean+Moorings

Estimates of data quantity 10Mo

Product delivery ftp from http://cdiac.ornl.gov/oceans/Moorings/Pacific.html or graphics from http://www.pmel.noaa.gov/co2/story/Open+Ocean+Moorings




Pricing Free

Access conditions To be checked with [email protected]



Miscellaneous

Comments Contact [email protected]

IS-09 NODC Product Name World Ocean Database 2009 No. Data type Vertical profile (CTD,XBT,..) of salinity and Temperature Source NODC Key Websites http://www.nodc.noaa.gov/OC5/WOD/pr_wod.html Version - Platform name -

Platform characteristics - Sensor(s) Sensor type Sensor key technical characteristics Analysis characteristics - References to technical specifications documents

-

Product format Ascii files Data gridding Data coverage: temporal Data coverage: spatial global Project Requirements Date required within project Subset in certain regions Use within project Relate in situ surface sss to vertical properties (stratification

mixed-layer depth, etc). Reason for selection THE historical database Subset or complete record needed subsets

Data quality Data calibration World Ocean Database (WOD) contains the World Ocean

Database 2009 (WOD09) with the full set of quality control used to create World Ocean Atlas 2009 (WOA09) and all updates to the database (Feb. 2009 to present) with only initial quality control

Data validation

Product limitations


Data availability

Available from

Availability time-scale -



Product delivery http://www.nodc.noaa.gov/OC5/WOD/pr_wod.html



Pricing Free

Access conditions



Miscellaneous

Comments

IS-10 ORE HYBAM Product Name ORE-HYBAM

No.

Data type River discharge times series

Source French Ministry of Higher education and Research

Key Websites http://www.ore-hybam.org/index.php/eng

Version -

Platform name -

Platform characteristics -

Sensor(s)

Sensor type

Sensor key technical characteristics

Analysis characteristics -


-

Product format Ascii/Netcdf files

Data gridding

Data coverage: temporal

Data coverage: spatial ORE HYBAM network stations (15 stations of the ORE HYBAM stricto sensu over the Amazon basin, a station on the Orinoco River and a station on the Congo River);


Date required within project Times series at Obidos (Amazon) and Bolivar (Orinoco) since 2010

Use within project Relate SMOS detected plume characteristics with discharge levels


Reason for selection

Subset or complete record needed subsets

Data quality

Data calibration

Data validation

Product limitations


Data availability

Available from 2003-noiw

Availability time-scale -daily

Estimates of data quantity Very good

Product delivery http://www.ore-hybam.org/index.php/eng/Data



Pricing Free

Access conditions



Miscellaneous

Comments

2.1.1 Model Output

NM-01 FOAM-Global FOAM-Global FOAM-Global No. Data type -Model

-Salinity (daily mean down to 20m, 3-hourly surface) -Temperature (daily mean down to 20m, 3-hourly surface) -Surface velocity (daily mean) -Mixed layer depth (daily mean) -Sea surface height (daily mean)

Source UK Met Office Key Websites http://www.ncof.co.uk/Deep-Ocean-Modelling.html Version Reanalysis Platform name N/A

Platform characteristics Ocean model data (from ¼ degree global NEMO). Sensor(s) N/A Sensor type N/A


Sensor key technical characteristics N/A

Analysis characteristics Model assimilates in situ T and S profile data (Argo, XBTs, CTDs, moored buoys), satellite SST data (GHRSST), in situ SST data (ships, drifting and moored buoys), altimeter SSH data and SSM/I sea-ice concentration data.


Storkey, D., E.W. Blockley, R. Furner, C. Giuavarc'h, D. Lea, M.J. Martin, R.M. Barciela, A. Hines, P. Hyder, J.R. Siddorn, 2010. Forecasting the ocean state using NEMO: The new FOAM system. J. Operational Oceanography, 3, 3-15.

Product format NETCDF Data gridding Tri-polar grid (ORCA025). This can be interpolated to a regular ¼

degree lat-long grid. Data coverage: temporal 1992-present. Data coverage: spatial Global Project Requirements Date required within project 01/01/10 to 31/12/2012 Use within project CS2, CS3, CS4, CS5 Reason for selection To understand how model and data compare, in preparation for

future assimilation. Subset or complete record needed 2010-2012

Data quality Data calibration Storkey, D., E.W. Blockley, R. Furner, C. Giuavarc'h, D. Lea, M.J.

Martin, R.M. Barciela, A. Hines, P. Hyder, J.R. Siddorn, 2010. Forecasting the ocean state using NEMO: The new FOAM system. J. Operational Oceanography, 3, 3-15.

Data validation Storkey, D., E.W. Blockley, R. Furner, C. Giuavarc'h, D. Lea, M.J. Martin, R.M. Barciela, A. Hines, P. Hyder, J.R. Siddorn, 2010. Forecasting the ocean state using NEMO: The new FOAM system. J. Operational Oceanography, 3, 3-15.

Product limitations Only sparse in situ salinity data from Argo currently assimilated.

Potential product upgrades Various developments on-going but unlikely to be available prior to use in this project.

Data availability

Available from UK Met Office

Availability time-scale Data available now. Extraction of required fields will take a couple of weeks.

Estimates of data quantity 150GB for 3 years of global data

Product delivery ftp from Met Office



Pricing Free

Access conditions Access for research use within the project.

Formal agreements with data suppliers Data suppliers are part of this project.

Third party redistribution. No redistribution of model data.

Miscellaneous

Comments None

NM-02 FOAM-N.Atl.


FOAM-NAtl FOAM_NAtl No. Data type -Model

-Salinity (daily mean down to 20m) -Temperature (daily mean down to 20m) -Surface velocity (daily mean) -Mixed layer depth (daily mean) -Sea surface height (daily mean)

Source UK Met Office Key Websites http://www.ncof.co.uk/Deep-Ocean-Modelling.html Version Operational Platform name N/A

Platform characteristics Ocean model data (from 1/12 degree North Atlantic NEMO). Sensor(s) N/A Sensor type N/A Sensor key technical characteristics N/A

Analysis characteristics Model assimilates in situ T and S profile data (Argo, XBTs, CTDs,

moored buoys), satellite SST data (GHRSST), in situ SST data (ships, drifting and moored buoys), altimeter SSH data and SSM/I sea-ice concentration data.


Storkey, D., E.W. Blockley, R. Furner, C. Giuavarc'h, D. Lea, M.J. Martin, R.M. Barciela, A. Hines, P. Hyder, J.R. Siddorn, 2010. Forecasting the ocean state using NEMO: The new FOAM system. J. Operational Oceanography, 3, 3-15.

Product format NETCDF Data gridding Rotated 1/12 degree lat/long grid This can be interpolated to an

unrotated 1/12 degree lat-long grid. Data coverage: temporal 15/03/2010-present Data coverage: spatial North Atlantic Project Requirements Date required within project 15/03/2010 to 31/12/2012 Use within project CS2 Reason for selection To understand how model and data compare, in preparation for

future assimilation. Subset or complete record needed 2010-2012

Data quality Data calibration Storkey, D., E.W. Blockley, R. Furner, C. Giuavarc'h, D. Lea, M.J.

Martin, R.M. Barciela, A. Hines, P. Hyder, J.R. Siddorn, 2010. Forecasting the ocean state using NEMO: The new FOAM system. J. Operational Oceanography, 3, 3-15.

Data validation Storkey, D., E.W. Blockley, R. Furner, C. Giuavarc'h, D. Lea, M.J. Martin, R.M. Barciela, A. Hines, P. Hyder, J.R. Siddorn, 2010. Forecasting the ocean state using NEMO: The new FOAM system. J. Operational Oceanography, 3, 3-15.

Product limitations Only sparse in situ salinity data from Argo currently assimilated.

Potential product upgrades Various developments on-going but unlikely to be available prior to use in this project.

Data availability

Available from UK Met Office

Availability time-scale Data available now. Extraction of required fields will take a couple of weeks.


Estimates of data quantity 60GB for 3 years of data

Product delivery ftp from Met Office



Pricing Free

Access conditions Access for research use within the project.

Formal agreements with data suppliers Data suppliers are part of this project.

Third party redistribution. No redistribution of model data.

Miscellaneous

Comments None

2.2 Data Design Justification The data sets selected for use in the project are the most complete sets of data, covering the project needs and publicly available today. Full use is being made of suitable ESA data. In very specific cases, when data are not publicly available, we also consider using data specificly made available to this project.

The data sets listed cover all aspects of the Case Studies described in the project SAP.

2.3 Risks and proposed solutions This section presents a table of risks relating to access and use of data sets listed in 2.1, e,g, any restrictive data ownership and use conditions, any NRT or future satellite data to be used.

Data risk Proposed solution

IS-02: These data are not yet in Coriolis data base but should be put before end 2013 (T. Carval, Coriolis, pers. Comm.)

We already got an agreement from Steve Reiser (PI of ARGO_STS) to access the original files.

IS-08: Only real time data is visible on web site after 2009

In case no verified data is available when we’ll start our study (in 2014) we’ll take contact with NOAA responsible and eventually conduct rough validation based on earlier measurements in the area. An alternative could be to focus into area covered by the SOCAT version 2 pCO2 data base recently released (http://www.socat.info/about.html).


3 Algorithm Theoretical Basis Description

This section is included to provide a summary overview of the Algorithms that will be implemented in the project as required by the SoW.

During the scientific analysis work of the project it is not planned to implement any new algorithms within the project. All case study scientific work will be based on existing data products. However this section has been updated for version 2.0 of the document to include further information on data and methods used in the Case Studies now that scientific work is well underway.

Case study 1: Amazon/Orinoco plumes Further details and references are included in:

Reul, N., et al. (2013), Sea Surface Salinity Observations from Space with the SMOS Satellite: A New Means to Monitor the Marine Branch of the Water Cycle, Surv Geophys, 1-42.

A range of satellite and in situ data sets are used in the present study with focus on the years 2010–2012 following the SMOS launch date. The data products are described below.

SMOS SSS Data SMOS (Soil Moisture and Ocean Salinity) is the European Space Agency (ESA)’s water mission (Kerr et al. 2010; Mecklenburg et al. 2012), an Earth Explorer Opportunity Mission approved under the Living Planet Program. SMOS was launched in November 2009, and the technical approach developed to achieve adequate radiometric accuracy, as well as spatial and temporal resolution compromising between land and ocean science requirements, is polarimetric interferometric radiometry (Ruf et al. 1988; Font et al. 2010) at L-band (frequency of *1.4 GHz). ESA produces so-called level 2 SSS, or L2 products, which correspond to instantaneous SSS retrievals under the satellite swath.

In the present study, level 2 SMOS SSS are from the first SMOS/ESA annual reprocessing campaign in which ESA level 1 v5.04 and level 2 v5.50 processors have been used. In these versions, significant improvements with respect to the flaws discovered in the first products (e.g., Reul et al. 2012) have been implemented (see a complete description in the Algorithm Theoretical Basis Document (ATBD) available at http://www.argans.co.uk/ smos/docs/deliverables/). Nevertheless, accuracy of these instantaneous SSS retrievals is rather low (*0.6–1.7), and space–time averaging of the level 2 products is needed (so-called level 3 SSS) to decrease the noise level in the retrievals.

Here, we used two types of composite SSS level 3 products generated in laboratories participating to the Expertise Center of the Centre Aval de Traitement des Donne ́es SMOS (CATDS, http://www.catds.fr), which is the French ground segment for the SMOS data. These products are built either from ESA level 1 products (Reul and Tenerelli 2011) or from ESA level 2 products (Boutin et al. 2012b).

These research products aim at assessing the quality of SMOS operational products (ESA level 2 and CATDS-OP level 3) and at studying new processing to be implemented in the future in operational chains. Main characteristics of these products are detailed in Table 1. CEC-IFREMER products have been used in Sects. 3, and 5, CEC-LOCEAN products in Sect. 4.


Overall accuracy of the 10-day composite products at 25-km resolution is on the order of 0.3 practical salinity scale in the tropical oceans (Reul and Tenerelli 2011).

Table 1 Summary of characteristics of CATDS-CEC SSS level 3 products:

CEC-Ifremer CEC-LOCEAN

SSS retrieval method SSS retrieved from first Stokes parameter (Reul and Tenerelli 2011)

SSS retrieved from polarized Tbs along dwell lines using an iterative retrieval (see ESA L2OS ATBD)

Region of the instrument field of view (FOV) considered for SSS retrieval

Alias free field of view only

Alias free field of view (AFFOV) and extended AFFOV along dwell lines with at least 130 Tb data samples in AFFOV (*±300 km from the swath center)

Tb filtering method Determined from interorbit consistency in incidence angles classes and thresholding

Determined from consistency along dwell lines as reported in ESA level 2 products

Galactic model Geometrical optics model

Kirchoff’s approx. scattering at 3 m/s

Roughness/foam models Empirical adjustment of Tb dependencies to wind speed

Empirical adjustment of parameters in roughness model and foam coverage models (Yin et al. 2012)

Calibration Single ocean target transformation (OTT) ? daily 5° 9 5° adjustment wrt World Ocean 2001 SSS climatology

Variable OTT (every 2 weeks synchronized with noise injection radiometer as defined in ESA reprocessing)

Average Simple average

Average weighted by theoretical error on retrieved SSS and spatial resolution

Ocean Surface Currents Here, we used the 1/3° resolution global surface current products from Ocean Surface Current Analyses Real time (OSCAR) (Bonjean and Lagerloef 2002; http://www.oscar. noaa.gov), directly calculated from satellite altimetry and ocean vector winds.

The OSCAR data processing system calculates sea surface velocities from satellite altimetry (AVISO), vector wind fields (QuikSCAT), as well as from sea surface temperature (Reynolds–Smith) using quasi-steady geostrophic, local wind-driven, and thermal wind dynamics. Near real-time velocities are calculated both on a 1° 9 1° and 1/3° 9 1/3° grid and on a *5-day time base over the global ocean. Surface currents are provided on the OSCAR Web site (http://www.oscar.noaa.gov) starting from 1992 along with validations with drifters and moorings. The 1/3° resolution is available for FTP download through ftp://esr.org/pub/datasets/SfcCurrents/ThirdDegree.

Rain, Evaporation and River Discharge Data


To estimate the rain rate over the oceans, we used three different satellite products. One is the monthly Tropical Rainfall Measuring Mission (TRMM) Composite Clima- tology (TCC) of surface precipitation based on 13 years of data from the TRMM. The TCC takes advantage of the information from multiple estimates of precipitation from TRMM to construct mean value maps over the tropics (36°N–36°S) for each month of the year at 0.5° latitude–longitude resolution. The first-time use of both active and passive microwave instruments on board TRMM has made it the foremost satellite for the study of precipitation in the tropics and has led to a better understanding of the underlying physics and distribution of precipitation in this region. The products are available at NASA Goddard

Space Flight Center Global Change Master Directory (http://gcmd.nasa.gov). The second type of satellite rain rate estimates that we used in the present study are the so- called ‘‘TRMM and Other Satellites’’ (3B42) products, obtained through the NASA/Giov- anni server (http://reason.gsfc.nasa.gov/OPS/Giovanni). The 3B42 estimates are 3 hourly at a spatial resolution of 0.25° with spatial extent covering a global belt (-180°W–180°E) extending from 50°S to 50°N latitude. The major inputs into the 3B42 algorithm are IR data from geostationary satellites and Passive Microwave data from the TRMM microwave imager (TMI), special sensor microwave imager (SSM/I), Advanced Microwave Sounding Unit (AMSU), and Advanced Microwave Sounding Radiometer-Earth Observing System (AMSR-E). The Special Sensor Microwave Imager (SSM/I) F16 and F17 orbits cross SMOS orbits within -20 min and ?40 min. Hence, numerous SMOS level 2 are collocated with SSMI rain rates (RR) within this range of time. In addition to the TRMM 3B42 products, we therefore used SSM/Is data sets to perform colocations between SMOS SSS and rain estimates. SSM/Is RR version 7 was used and downloaded from http://www.remss.com.

The evaporation (E) data set was taken from the version 3 products of the Objectively Analyzed air-sea Fluxes (OAFlux) project (Yu and Weller 2007).

Finally, the discharge data for the Amazon, Orinoco, and Congo rivers were obtained from the Environmental Research Observatory HYBAM (geodynamical, hydrological, and biogeochemical control of erosion/alteration and material transport in the Amazon basin) Web site (http://www.ore-hybam.org/).

Ocean Color Products To study the spatiotemporal coherency between SSS signals from some major tropical river plumes and ocean color properties, we used the level 3 daily, 4-km resolution estimates of the absorption coefficient of colored detrital matter (CDM) at 443 nm. These products processed and distributed by ACRI-ST GlobColour service are supported by the EU FP7 MyOcean2 and the ESA GlobColour Projects, using ESA ENVISAT MERIS data, NASA MODIS and SeaWiFS data. These products have been averaged at the SMOS L3 product 0.25° resolution, with a 10-day running mean.

In Situ Data Salinity measurements from Argo floats are provided by the Coriolis data center (http:// www.coriolis.eu.org/). The upper ocean salinity values recorded between 4- and 10-m depth will be referred to as Argo SSS following Boutin et al. (2012b).

Global SSS maps are derived from delayed time quality checked in situ measurements (Argo and ship) by IFREMER/LPO, Laboratoire de Physique des Oceans, using the In Situ Analysis System (ISAS) optimal interpolation (D7CA2S0 re-analysis product) (see a method description on http://wwz.ifremer.fr/lpo/SO-Argo-France/Products/Global-Ocean- T-S/Monthly-fields-2004-2010 and in (Gaillard et al. 2009)). The choice for the time and space scales used in that method results from a compromise between what is known of ocean time and space scales and what can actually be resolved with the Argo array (3°, 10 days); two length scales are considered: the first one is isotropic and equal to 300 km, the second one is set equal to 4 times the average Rossby radius of deformation of the area. As a result, we


expect these maps being smoother, especially in tropical areas, than SMOS SSS maps averaged over 0.25° 9 0.25° or 1° 9 1°.

Case study 2: Agulhas, Gulf Stream Further details and references are included in:

Reul, N., B. Chapron, T. Lee, C. Donlon, J. Boutin, and G. Alory (2014), Sea surface salinity structure of the meandering Gulf Stream revealed by SMOS sensor, Geophysical Research Letters, 2014GL059215.

SMOS is a polar-orbiting satellite carrying an interferometric radiometer operated at 1.4 GHz and covering the entire globe with a 3 day repeat subcycle. SSS is retrieved from the raw brightness temperature data across a swath of ~1000 km with a spatial resolution of 35 to 50 km. Swath products exhibit a global mean error of 0.52 practical salinity units (psu), decreasing to about 0.3 psu in the tropical oceans [Mecklenburg et al., 2012]. For this study, we used the Centre Aval de Traitement des Données SMOS (CATDS, www.catds.fr) Expertise Center-Ocean Salinity SMOS SSS (IFREMER V02) products [Reul and Ifremer CATDS-CECOS Team, 2011]. radio frequency interference (RFI) from military radar arrays installed over North America heavily contaminated SMOS data in the western North Atlantic area during the first 2 years of mission operations (2010–2011). Since the end of 2011, ongoing actions to refurbish L band radar stations in Canada [Daganzo-Eusebio et al., 2012] have led to a dramatic reduction in RFI contamination. Consequently, we only consider SMOS data acquired during 2012 for this study. Data in 2012 were first processed to provide a level 3 daily mean gridded SSS field at a resolution of 0.25° × 0.25° for the complete year. Composite products were then generated using a running mean 11 days, 0.5° window.

The accuracy of SMOS SSS in the study region is assessed (see supporting information section S1) by comparing satellite products to in situ underway thermosalinograph (TSG) data and salinity measurements derived from Argo floats in the upper 10 m of the ocean. In situ data collected over the spatial domain (77°W–40°W; 30°N–50°N) were colocalized with SMOS 11 day products for 2012. Differences between in situ and satellite SSS observations exhibit a standard deviation of ~0.5 psu. While the physical explanation remains unclear, SMOS data quality is found to degrade as SST drops below ~13°C, with an increasing bias (SMOS SSS data being saltier than in situ) from ~0.5 psu between 5°C and 13°C to more than 1 psu below 5°C. Measurements from Argo floats are provided by the Coriolis data center (http://www.coriolis.eu.org/); Argo-based climatologies were derived over the period 2004–2010 [Kolodziejczyk and Gaillard, 2012]. Underway TSG data from Voluntary Observing Ships and Research Vessels were accessed through the Global Ocean Surface Underway Data (http://www.gosud.org/) and Shipboard Automated Meteorological and Oceanographic System (http://samos.coaps.fsu.edu/) data centers.

We used the 1/3° resolution surface current products from Ocean Surface Current Analyses Realtime (OSCAR) [Bonjean and Lagerloef, 2002] (http://www.oscar.noaa.gov/), directly calculated from satellite altimetry and ocean vector winds. At the mesoscale, surface currents are dominated by geostrophic flow estimated from altimetry. Standard altimetry absolute dynamic topography (SSH) and anomaly (sea level anomaly) gridded data are provided by AVISO (Archiving, Validation, and Interpretationof Satellite Oceanographic data, http://www.aviso.oceanobs.com/).

Satellite SST data used here are the Group for High-Resolution SST (GHRSST) [Donlon et al., 2007] level 4 products generated by the ODYSSEA processing chain at Ifremer [Piolle et al., 2010]. In addition, 8 day composite of chlorophyll concentration from Aqua/Moderate


Resolution Imaging Spectroradiometer (MODIS) were obtained from NASA Goddard Space Flight Center [O’Reilly et al., 2000].

Case study 3: Tropical Pacific & Atlantic Further details and references are included in:

Boutin, J., N. Martin, G. Reverdin, S. Morisset, X. Yin, L. Centurioni, N. Reul, Sea Surface Salinity under rain cells: SMOS satellite and in-situ drifters observations, Journal Geophys. Res.-Oceans (in revision).

SMOS The SMOS mission [Kerr et al., 2010] has been launched in November 2009, on a sun-synchronous circular orbit with a local equator crossing time at 6 AM on ascending node. It carries a L-band interferometric radiometer. This new technology allows reconstructing bi-dimensional multi-angular images of Tb with a mean spatial resolution of 43km. Individual measurements are very noisy (the typical noise on individual retrieved S1cm is 0.6 in tropical regions) ; however this noise can be reduced by averaging S1cm in space and time [Boutin et al., 2004]. The retrieval scheme implemented in the ESA (European Space Agency) processing retrieves SMOS S1cm, wind speed, sea surface temperature (SST), total electron content, and their theoretical errors, from the multiangular and polarised SMOS Tbs collected at an earth pixel during the satellite pass, using Levenberg-Marquard (L.M.) minimization method as described in [Zine et al., 2008]. Prior values for wind speed and SST are taken from European Centre for Medium-Range Weather Forecasts (ECMWF); in the ESA operational chain, errors of 2 m s-1 and of 1°C have been attributed to wind components and SST respectively. The theoretical errors are retrieved from the Jacobian of Tb with respect to the geophysical parameters and from the a posteriori covariance matrix of errors in Tb and geophysical parameters (see [Zine et al., 2008]). At first order, the theoretical error of S1cm depends on the number of Tb data used in the retrieval and on SST (because of the strong dependency of dTb/dSSS with SST).

In the present study, we use the level 2 SMOS S1cm from the first SMOS/ESA annual reprocessing campaign in which ESA level 1 v5.04 and level 2 v5.50 processors have been used (see a complete description in the Algorithm Theoretical Basis Document (ATBD) available on http://www.argans.co.uk/smos/docs/deliverables/delivered/ATBD/SO-TN-ARG-GS-0007_L2OS-ATBD_v3.8_111117.pdf). Large scale seasonal biases, likely due to flaws of the thermal antenna model [Kainulainen et al., 2012] are still present in this version. In the southern tropical Pacific Ocean, they are corrected by the application of the Ocean Target Transformation so that the rms error of monthly – 100x100km2 SMOS S1cm with respect to ship Sbulk has been found equal to 0.20 in the south east Pacific between 0° and 30°S, a region with very few rain events [Hasson et al., 2013]. On another hand, [Hernandez et al., 2014] have shown large biases (several tenths of pss) in the northern subtropical Atlantic region (15°N-35°N), largest in boreal winter. Once these large scale monthly biases are removed, the rmse of monthly – 100x100km2 SMOS S1cm with respect to ship measurements is equal to 0.15. We have chosen the boreal Summer season for collocating SMOS SSS with ARGO SSS in the ITCZ (Intertropical Convergence Zone) because it corresponds to a period of relatively low biases north of the equator as found in the North Atlantic subtropical regions [Boutin et al., 2013]. Hence in the present study, we extend the SMOS S1cm-ARGO Sbulk comparison in the ITCZ region to boreal summers 2010 and 2012; a much longer period (June 2010 to February 2011) is considered for the SMOS-ARGO comparison in the SPCZ (south of the equator) for which the OTT is assumed to correct for seasonal biases.


We use ESA level 2 SMOS S1cm retrieved with model 1, which makes use of the [Yin et al., 2012b] roughness model; only ascending orbits are considered in order to minimize uncertainties linked to Faraday rotation and to diurnal SST cycle.

The filtering of the ESA level 2 SMOS S1cm is performed as follows: we retain grid points flagged as valid, as well as with successful retrieval, with a good fit between measured and modelled Tbs (tests on Chi2 and Chi2_P as defined in ATBD), with less than 20 iterations of the Levenberg and Marquardt retrieval process, with no suspicious ice or numerous outliers. In addition, in order to 1) avoid too noisy retrievals at the edge of the swath and 2) inaccuracy due to lower accuracy of ECMWF forecasts or of the roughness model at very low and high wind speed, we only consider SMOS S1cm retrieved in grid points with 1) more than 130 Tb coming from the alias free field of view region (roughly corresponding to S1cm retrieved at +/-300km from the centre of the track) and 2) ECMWF wind speed between 3 and 12m/s. The averages of ESA level 2 SMOS S1cm are weighted with theoretical error and measurement resolution as described in [Yin et al., 2012a]. Only averages made with more than 30 individual SSS are retained. With these criteria, a grid point is seen approximately once every 5 days, during ascending orbits.

Two step retrieval algorithm [Yin et al., 2013] have shown that in cases when ECMWF wind speed differs from SSM/I radiometric wind speed, the SMOS retrieval scheme corrects part of the difference between these two wind speeds and this improves the quality of the retrieved S1cm. In cases of large differences between these two wind speeds, they tested a SMOS retrieval with a larger a priori error on wind speed (5m s-1 instead of 2m s-1). This resulted in a retrieved wind speed closer to SSM/I, a smaller bias on SSS, but also increased noise on retrieved parameters. In order to correct biases without increasing too much the noise, we have developed an alternative retrieval algorithm (two step algorithm). In a first step, the error on a priori ECWMF wind speed is set to 5m s-1. This results in a very noisy retrieved wind speed over the 15km resolution ISEA grid which is then filtered using a bidimentional spatial median filtering having a 50km radius. In a second step, the smoothed retrieved wind speed is used as a priori wind speed (instead of ECMWF) with an error set to be 2m s-1. This method has been successfully tested in the eastern equatorial Pacific region (5°S-1°N; 90°W-130°W) in August 2010 where a systematic bias is observed on both SMOS retrieved S1cm (a 0.46 difference with respect to ARGO Sbulk) and on ECMWF wind speed (1.4m s-1 difference with respect to radiometric SSM/I wind speed), although the operational method already corrects for half of the ECMWF minus SSM/I wind speed difference [Yin et al., 2013]. When applying the two step retrieval method instead of the operational retrieval method, the SMOS S1cm bias with respect to ARGO Sbulk is reduced from 0.46 to 0.26, and the SMOS retrieved wind speed is decreased by 0.4m s-1 making it very close (0.3m s-1 difference) to the wind speed retrieved when using SSM/I wind speed as the guess instead of ECMWF wind speed (Yin et al., poster at the ESA Living Planet Symposium, http://www.argans.co.uk/smos/pages/posters.php?poster=LPS2013_Yinetal.pdf).

Rain splash modifies the ocean roughness as seen by the radiometer, which should be affecting more Th than Tv at large incidence angle contrary to what is induced by a change in S1cm. Hence, given that SMOS retrieval uses the polarized Tb at various incidence angles to separate SSS and roughness (parametrized in terms of wind speed) signals, one expects the retrieved wind speed to be modified by a change in rain induced roughness. In order to test the importance of this change for rainy SMOS measurements, we have looked at S1cm retrieved with the two step algorithm.

Natural variability of SMOS SSS Individual SMOS SSS are very noisy. In order to distinguish between the expected noise due to the radiometric noise and the variability due to other effects, we compute the quadratic difference between the standard deviation of SMOS SSS within one month and 100x100km2


and the theoretical error averaged over the same time and spatial scales using weights depending on the SMOS measurements resolution and on the theoretical errors (see Equation A 8 of [Yin et al., 2012a]). Assuming that the theoretical error provided with retrieved SMOS SSS is a realistic estimate that takes into account all error sources, this computation represents the natural variability in one month of S1cm averaged over 43 km, as in SMOS.

Satellite Rain Rate and Wind Speed Satellite rain rates (RR) and wind speeds from WindSat version 7.0.1, SSM/Is F15, F16 and F17 version 7, AMSR-E and TMI version 4, distributed by Remote Sensing System (www.remss.com) have been used. They were derived from the Unified Microwave Ocean Retrieval Algorithm (UMORA) described in [Hilburn and Wentz, 2008] after thorough intercalibration of the SSM/Is data as described in [Wentz, 2013]. SSMI F15 quality degrades after August 2006 and RemSS recommend not to use it for climate studies. Nevertheless, we consider it in our RR colocations as we are not performing climate studies but we are considering large instantaneous RR variability for which SSM/I F15 can be very complementary to the other satellite measuring RR. For most SSM/I missions, the local equator crossing time is close to 6 PM, although for some of them (in particular SSM/I F15), it drifts in time (see http://www.remss.com/support/crossing-times). As a consequence, the closest colocations between SMOS and satellite RR are found with SSM/I. In July-September 2010, the period during which the analysis of the SMOS S1cm-ARGO Sbulk differences was the most extensive, the majority of SSM/I measurements were at more than 30mn from SMOS measurements, although by less than 1:30. Nevertheless, given the SSM/I time drift, in July-September 2012, the majority of SSM/I F16 and SSM/I F17 were closer to SMOS measurements, within [-30mn; +15mn]. As a consequence, we consider two time intervals in our SMOS-SSM/I matchups: [-60mn; +30mn] and [-30mn; +15mn]; only the satellite RR closest in time with SMOS SSS is retained.

In addition to individual satellite RR products, in order to get information about RR whatever the local time is, we use the TRMM3B42 product version 7 (http://disc.sci.gsfc.nasa.gov/precipitation/documentation/TRMM_README/TRMM_3B42_readme.shtml) which provides RR estimate over 3 hours. It is used to distinguish between ARGO SSS measured under rainy or non rainy conditions. However, the 3-hour time resolution is insufficient to characterize the correlation between SMOS S1cm – ARGO Sbulk and RR: using the TRM3B42 RR product instead of SSM/I RR colocated within (-60mn;+30mn) degrades the correlation coefficients by a factor 1.4 to 1.6.

ARGO SSS We use measurements from ARGO floats provided by the Coriolis data centre (http://www.coriolis.eu.org/), with a quality flag equal to 1, in agreement with real time quality checks and, for delayed time data, with statistical consistency checks [Carval et al., 2012]. In order to avoid unpumped measurements (see [Boutin et al., 2013]) we use the closest ARGO salinity to the sea surface, provided it is measured between 4m and 10m depth, without any interpolation to the surface. We will later refer to this measurement as ARGO Sbulk. These ARGO Sbulk are colocated with SMOS S1cm within a radius of +/-5days and +/-50km. Contrary to what was done in [Boutin et al., 2013] in which we colocated all ARGO Sbulk with SMOS S1cm whatever the rain conditions were at the time of ARGO measurement, in the present study we exclude rainy ARGO Sbulk identified by TRMM3B42 product within -2h and +1h from each ARGO measurement. This test identifies that 24% of the ARGO measurements have occured at less than 2 hours from a rain event in the ITCZ region. The ARGO Sbulk taken under rainy conditions are discarded from the SMOS S1cm – ARGO Sbulk. Thus, S1cm minus Sbulk will represent an upper bound of the vertical stratification effect.


SVP drifters A large set of Surface Velocity Drifters (SVP) measuring conductivity and temperature at about 45cm depth have been deployed for the new salinity satellite calibration and validation. They have been thoroughly quality checked [Reverdin et al., 2014]. They show large salinity variability often associated with rainfall [Reverdin et al., 2012]. Hence we develop a method that automatically detects sharp and local decrease of SSS, SSSmin, possibly affected by rain. A S45cm_min is identified as affected by rainfall if:

- the difference between the median of the S45cm measured every 30mn during the 6 hours preceeding S45cm_min (S45cm_ref) and S45cm_min is larger than 0.4,

- the difference between the median of the S45cm measured every 30mn during the 6 hours after S45cm_min and S45cm_min is larger than 0.2,

- S45cm_min is a local minimum : at least 0.02 smaller than S45cm measured just before and just after S45cm_min and S45cm measured every half hour during the 2.5hours after S45cm_min must increase by at least 0.01 per 0.5hr, in order to avoid misidentifying crossing of fronts as rainfall events.

Using this test, 470 events of sharp and large S45cm decrease events have been identified since 2009. This number was not sufficient to obtain reliable comparisons between drifters and SMOS S1cm under rainy conditions, especially since such colocations must be done within a small temporal radius and because the large noise on SMOS S1cm requires to average a large number of measurements to get statistically significant results.

Hence, instead of a direct comparison of drifter S45cm with SMOS S1cm under rain cells, we compare the SSS decrease associated with satellite rain rate either deduced from SMOS S1cm or from drifters S45cm.

SMOS SSS decrease associated with Rain Rates: Two methods have been tested to estimate the SMOS S1cm decrease under rain cells. The first method, similar to the one used in [Boutin et al., 2013], is based on differences between SMOS S1cm and ARGO Sbulk by taking SMOS S1cm at +/-50km and +/-5days from ARGO floats. Given the intermittency of rain, we do not average SMOS measurements. Instead, the SMOS S1cm - ARGO Sbulk differences are analyzed as a function of the SSM/I satellite rain rate acquired the closest in time to the SMOS SSS measurement within an interval of either -60mn and +30mn, or -30mn and +15mn. This method was applied in the ITCZ (5°N-15°N; 180°W-110°W) as in [Boutin et al., 2013], and in the SPCZ (18°S-2°S; 160°E-170°W). As mentioned earlier, the ITCZ study is done in boreal summer to minimize SMOS large scale biases effects. On the other hand, since large scale biases are expected to be small in the latitudinal range of the SPCZ region and given the smaller size of the chosen ‘SPCZ’ region, we extend the studied period to June 2010 to March 2011 in order to get a more significant number of SMOS/ARGO-RR colocations.

The second method, independent of any in situ SSS comparison, correlates the spatial variability of SMOS S1cm with the one in RR maps.

Spatial variability of SMOS S1cm associated with the presence of rain cell has been determined from a comparison with a spatial field of satellite RR taken as close as possible in time from the SMOS SSS field (at typically less than half an hour), as follows. First, we identify, over a given region, the SMOS pixels located at less than 100km from a pixel with a satellite RR larger than 5mm hr-1. For each of these rainy SMOS S1cm pixels, we estimate the rain effect on S1cm as the difference between the local SMOS S1cm and an estimated ‘rain-free’ S1cm taken as the mean of the S1cm colocated with a null RR within less than 150km from the local SMOS S1cm.


Drifters and RR colocations Amongst the 470 sharp S45cm decreases observed by drifters and identified as described in section II.4, 24 have been colocated with satellite RR passes at less than +/-15mn; the +/-15mn temporal radius corresponds to half the interval between successive drifter S45cm measurements while a huge variability is observed on successive drifter S45cm around the S45cm minimum, likely due to the temporal variability of precipitation. The magnitude of the S45cm decrease, DSSS, has been estimated as S45cm_min - S45cm_ref (see section II.4) in most cases, except if the decrease appears to be discontinuous: in that case S45cm_ref is taken equal to the local maximum preceding S45cm_min. We associate DSSS with the average of the rain rates in 0.25° pixels which centers are at less than 50km from the drifter location at the time of DSSS; this corresponds approximately to averaging RR measured in the pixel containing the drifters, and in the ones adjacent to it. We do not only consider RR in the pixel containing the drifter in order to smooth the large temporal variability of rain within 15mn; actually successive SSM/I maps, at less than 15mn interval, show a huge temporal variability; when smoothing this variability over 9 RR pixels, it is much reduced.

It is very difficult to get information about wind speed under rain cells. Actually most of satellite wind speeds are flagged under rain conditions; an ‘All Weather’ [Meissner and Wentz, 2009] product containing a wind speed under rain cells is provided only with WindSat, while only 3 matchups have been found with WindSat. Hence, for the other matchups, we very crudely estimate a range of wind speed from a visual inspection of the radiometric wind speed maps around the rain cell. We classify our matchups in three categories, very low wind speed (less than 3m s-1), moderate wind speed (3-12m s-1) and strong wind speed (larger than 15m s-1; there were no matchups corresponding to a wind speed between 12 and 15m s-1).

Case study 4: Sub-tropical North Atlantic (SPURS) Further details and references are included in:

O. Hernandez, J. Boutin, N. Kolodziejczyk, G. Reverdin, N. Martin, F.Gaillard, N. Reul, and J.L. Vergely, SMOS salinity in the subtropical north Atlantic salinity maximum: Part I: Comparison with Aquarius and in situ salinity, accepted with minor revisions, Journal Geophys. Res.-Oceans, 2014.

In situ data In situ Analysis System (ISAS) SSS

In this study, we use monthly fields of salinity obtained with ISAS (In Situ Analysis System), an optimal estimation tool designed for the synthesis of the Argo global dataset [Gaillard et al., 2009]. We use the version V6 D7CA2S0 [Gaillard, 2012], which covers the period 2004-2012, with a time overlap with SMOS measurements over the 2010-2012 period. We also used the monthly ISAS climatology (data from 2004 to 2012, version GLA13) [Kolodziejczyk and Gaillard, 2012].

The interpolated fields were produced over the global ocean by the ISAS project with datasets downloaded from the Coriolis data center. The field of analysis has a global coverage within 70°S-70°N.

The optimal interpolation is computed over a ½ degree grid and involves a structure function modeled as the sum of two Gaussian functions, each associated with specific time and space scales, resulting in a smoothing over typically 3 degrees (for details see Gaillard et al., 2009). In our study, these fields have been interpolated linearly onto a regular 0.25 degree grid. This interpolation is not expected to bring any new information with respect to the


original ISAS field about small scale structures and is only intended to facilitate the comparisons with other fields produced at this spatial scale.

The temperature and salinity fields are reconstructed on 152 levels ranging from 0 to 2000 m depth. We used the first level, that mostly referred to the shallowest valid Argo measurement between the surface and about 5 m depth. The major contribution to ISAS are the profiling floats from the Argo array. This data subset is complemented by data from the Tropical Moored Buoy Array program (TAO/TRITON (Tropical Atmosphere Ocean/Triangle Trans-Ocean Buoy Network), PIRATA (Prediction and Research Moored Array in the Atlantic), RAMA (Research Moored Array for African-Asian-Australian Monsoon Analysis and Prediction)) mooring array. A few CTD profiles transmitted in real time are used but XBTs and X-CTDs were excluded from the analysis because of uncertainties in the fall rate.

SSS ship In situ near surface salinity is provided by thermosalinographs (TSG) mounted on the merchant vessels Toucan and Colibri that cross the North Atlantic subtropical sea surface salinity maximum (SSM). These measurements onboard ships of opportunity are a contribution to French Sea Surface Salinity Observation Service (http://www.legos.obs.mip.fr/observations/sss). 39 transects with salinity data have been collected between 2010 and 2012 from Western Europe to northern South America. The near surface sea water is pumped on the side of the immersed ship's hull at about 5 meters depth. The nominal horizontal resolution is about 2.5 km. The typical duration of ship transect between Europe and South America is 10 days for each ship. Data are systematically post-calibrated with water samples and Argo data when they are available. Only data with "Adjusted" and "Good" or "Probably Good" flags data are used in this study (http://www.legos.obs-mip.fr/observations/sss/datadelivery/dmdata). The typical error on these TSG measurements is 0.01-0.02; on the crossings used in our study the difference between the calibrated TSG salinities and the water sample salinities vary between 0.01 and 0.08, part of this difference being possibly due to errors in the water sample salinities.

TSG data were also available from the research vessel (RV) Thalassa during the STRASSE (SubTRopical Atlantic Surface Salinity Experiment) cruise, from the RV Knorr during the SPURS-1 (Salinity Processes in the Upper Ocean Regional Study) cruise, from the RV Discovery (Di382) during the RAPID cruise (McCarthy et al. 2012) and from the RRS James Cook (JC079) during the AMT 22 (Atlantic Meridional Transect) cruise. Salinity of these cruise data has been validated with errors on the order of 0.005. We adopt the practical salinity scale (pss-78), defining salinity as a conductivity ratio, which does not have physical units.

Satellite data SMOS data SMOS satellite was launched in November 2009 into a sun-synchronous orbit at 758 km crossing the equator twice a day at 6 a.m. in ascending phase and at 6 p.m. in descending phase [Mecklenburg et al., 2012]. The SMOS mission carries the L-band (1.4 GHz) Microwave Imaging.

Radiometer with Aperture Synthesis (MIRAS) instrument from which a bi-dimensional field of view (FOV) of brightness temperatures (Tb) at various incidence angles is reconstructed. SMOS SSS considered in this study are based on Tb measurements at less than 300 km from the satellite center track. Coverage of the global ocean is achieved every 3 days with a repeat cycle of 149 days (sub-cycle of 18 days) and a nominal spatial resolution of 43 km on average over the FOV.

We first use the SMOS ESA L2 SSS, reprocessed using ESA version 5 processors. In this reprocessing, an 'Ocean Target Transformation' (OTT) correction was applied every two


weeks. This correction aims at correcting SMOS brightness temperatures (Tb) from systematic differences with respect to modelled Tb in the FOV [Yin et al., 2012]. It is computed from a large oceanic region far from land and a Radio Frequency Interference (RFI) contamination between 45°S and 5°S in the eastern Pacific [Yin et al., 2013]. However, seasonal and latitudinal biases are still present. Attempts to reduce such biases (including improved SMOS raw data calibration as well as the use of a time-varying OTT) remains under study [Martin-Neira et al., 2013; Yin et al., 2013].

SMOS SSS and their theoretical errors are retrieved from multi-incidence brightness temperatures (Tbs) collected at an earth pixel during the satellite pass, using Levenberg-Marquard (L.M.) minimization method as described in [Zine et al., 2008], after adjusting direct models with SMOS measurements (see a complete description in the Algorithm Theoretical Basis Document (ATBD) available at: http://www.argans.co.uk). The theoretical error is retrieved from the Jacobian of Tb with respect to the geophysical parameters and from the a posteriori covariance matrix of errors in Tb and geophysical parameters (see [Zine et al., 2008]). To first order, the theoretical error depends on the number of Tb data used in the SSS retrieval and on Sea Surface Temperature (SST) (because of the strong dependency of dTb/dSSS with SST). Relative accuracy of these SMOS data has been estimated as ~0.3-0.5 or better in tropical and subptropical regions for SSS averaged over 100 x 100 km2 and 10 days [Boutin et al., 2013].

The Level 3 product named SMOS CATDS CEC LOCEAN_v2013 has been generated from the above ESA Level 2 v5 reprocessed products (L2OS v5 wind-model 1) using only retrievals performed under moderate wind speed (3-12 ms-1 and according to the flags described in Boutin et al. [2013], except that the galactic noise flag was not tested (data affected by large galactic noise are nevertheless sorted out), and land mask is only 40 km. Level 3 SSS are flagged and averaged over one month, 100 x 100 km2 and oversampled every 0.25° When averaging Level 2 SSS, each retrieved SSS is weighted by its spatial resolution and its theoretical uncertainty as derived by the L.M. algorithm (for more details see Yin et al. [2012]).

Two other LOCEAN products were built for the purpose of this study. First, a SMOS LOCEAN 0.25° product is averaged following the same method as SMOS CATDS CEC LOCEAN maps, but the average is performed over 0.25°x0.25°instead of 100 x 100 km2. Second, a SMOS LOCEAN OI is produced from SMOS LOCEAN 0.5° product optimally interpolated with a similar method and same spatial correlation lengths as the one applied by ISAS [see Gaillard et al., 2009].

Apart from the Level 2 iterative L.M. retrieval, the IFREMER expertise center of CATDS (Centre Aval de Traitement des donnÈes SMOS, www.catds.fr) has developped an alternative processing chain starting from Level 1B products, in which the retrieval is simpler (SSS is retrieved from the first Stokes parameter, wind speed is not retrieved), RFI filtering is more efficient, only one OTT is applied over the whole period and a large scale bias correction is applied. We use monthly SMOS-CATDS CEC IFREMER SSS maps, averaged over 50 x 50 km2 and oversampled every 0.25° with a daily 5°x 5° adjustment with respect to World Ocean 2001 climatology. In this product, RFI percentage is estimated at each pixel. Only pixels for which the RFI percentage is null are used in this study. Nevertheless, undetected small amplitude residual contamination possibly remain in the data.

Our study focuses on the period July 2010 to December 2012. In December 2010, the last week of the month was not sampled due to an electrical stability test aboard the SMOS satellite. In January 2011, the 3 first weeks were degraded due to problems on one satellite arm (http://earth.eo.esa.int/missions/smos/available\_data\_processing.html). For these reasons, December 2010 and January 2011 are excluded from the analysis.


Aquarius data The Aquarius satellite was launched in June 2011 into a polar sun-synchronous orbit at 657 km crossing the equator twice a day at 6 p.m. in ascending phase (Orbit A) and at 6 a.m. in descending phase (Orbit D). It carries out a microwave radiometer at 1.413 GHz along with a scatterometer at 1.26 GHz for surface roughness correction. The nominal resolution of Aquarius satellite is about 100 km with a 7 days global coverage.

As for SMOS, several Level 3 products are available depending on the processing made. Here we use the two latest Level 3 versions of Aquarius data released to the scientific community (http://podaac.jpl.nasa.gov/datasetlist?search=aquarius): the Aquarius version 2.0 (see Aquarius Algorithm Theoretical Basis Documents (ATBD), [Wentz and Le Vine, 2012; Lagerloef et al., 2013]) and the Aquarius CAP version 2.0 [Yueh et al., 2012]. For both versions, we use the monthly spatial maps at 1° spatial resolution. In the Aquarius V2 CAP version, SSS is retrieved with the Combined Active Passive (CAP) algorithm which utilizes simultaneously data from the on board radiometer and scatterometer to retrieve SSS, wind speed and direction by minimizing the sum of squared differences between model and observations. We have performed some tests with the temporary version V2.5.1 which is precursor of the new version 3.0 that should be released in Spring 2014. Although large scale biases were reduced, we could not evidence improvement in the detection of mesoscale features, so that we prefer to keep the V2.0 official version.

Methods Definition of the region under study The goal of our study is to evaluate SMOS and Aquarius performances over the open ocean. Hence, we first conduct a preliminary study to avoid regions strongly contaminated by continent vicinity and by RFI. We define our region based on these criteria. Land contamination varies depending on the location of the pixel across track [Vergely et al., 2013]. Given that the orbit of SMOS is not repetitive over one month, land contamination is expected to generate artificial SSS variability, in addition to the already observed bias.

RFI sources vary in time both because the RFI emissions signals vary and because the contamination will depend on the location of the RFI contaminated point in the SMOS FOV. Hence, these contaminations are expected to artificially increase the SMOS brightness temperature variability, and hence the SSS variability observed within one month.

We estimate the variability of SMOS SSS retrieved along swath at about 40 km resolution within 100 x 100 km2 and one month from the standard deviation (σ) of SMOS L2 SSS. We define the natural variability of the SSS Evar in SMOS as:

Evar = !√(σ2 − Eth2)

where σ2 is the total variance of SMOS SSS, including natural variability, variability due to pollution sources (RFI, land contamination) and the theoretical error Eth of the retrieved SSS related to radiometric noise and uncertainties of auxiliary parameters used in the retrieval. Thus, in case of no external pollution sources, we expect that Evar is the natural variability retrieved from SMOS SSS.

The quadratic means of σ and Evar were calculated over the period July 2010 to December 2012. σ is higher than 0.6 everywhere, a large part comes from the radiometric noise: Eth in this region ranges from 0.5 to 0.8. On ship SSS, we always observe natural variability along 100 km to be lower than 0.4. Unrealistically high values of Evar are observed close to continents and in the northern region where RFI is expected. Therefore, we decide to bound the region under study to 50°W to 27°W in longitude and to 15°N to 35°N in latitude. This region will be referred to as the North Subtropical Atlantic (NSA).


In January 2012, we observe large anomalies between quadratic means of σ and expected theoretical error aligned with ascending orbits passes. By analysing individual orbits, we observe abnormally low SMOS SSS (as low as 31) along ascending orbits from January 26 - 29 crossing the eastern part of our region. These values were not filtered by SMOS flags but are likely due to RFI; hence in January, only the period of January 1-26 was retained in the monthly average.

Bias corrections Large scale comparisons of regional averaged SSS maps show that the SSS seasonal variability in the NSA region calculated from SMOS and Aquarius is not consistent with observations. SMOS CEC LOCEAN and Aquarius do not consistently reproduce the seasonal variability. SMOS CEC LOCEAN presents a very strong boreal winter bias (up to -0.4) that is visible each year, possibly due to strong sun contamination on descending orbits at that time. The SMOS CEC IFREMER reproduces the observed seasonal variability with ISAS products. However, this feature is expected because the IFREMER product is adjusted to the World Ocean Climatology (WOA 2001, [Conkright and Boyer, 2002]) with a daily 5°x 5° adjustment. Tests performed with Aquarius version 2.5.1 which is precursor of the future version 3 to be released soon, indicate that the large scale seasonal biases are much reduced especially with CAPV2.5.1.

The ISAS SSS products between 2010-2012 is saltier (about 0.1) than the climatological SSS seasonal cycle derived from WOA 2009. This is likely an effect of the long term trend of increasing salinity in North Subtropical Atlantic during the last decade year [Reverdin et al., 2007; Gordon and Giulivi, 2008; Durack et al., 2010; Terray et al., 2012].

In order to study the spatial variability of SSS, we first correct both satellite datasets (SMOS and Aquarius) from a bias B1 (see below) with respect to ISAS, estimated each month from an average in the NSA region.

B1 =< SSSSAT >NSA − < SSSISAS >NSA

where < SSSSAT >NSA is the satellite SSS averaged over the NSA region and < SSSISAS >NSA is the ISAS SSS averaged over the NSA region.

Colocations and statistics Precise collocations between in situ data and SMOS data are performed by averaging Level 2 SMOS SSS at ±50 km and ±9 days, using the same flags as described in Boutin et al. [2013]. The ±9 days are chosen to cover the 18-day SMOS repeat sub-cycle.

Statistics of the differences between satellite and in situ SSS were derived: mean bias error (MBE), the root mean square error (RMSE), the correlation coefficient (r) and the least square fit. All correlation coefficients that we report are significantly nonzero at 99% interval confidence. To analyse the significance of the difference in correlation coefficients between the products, we use the fisher r-to-z transformation [Zar, 2010]. We also calculate the standard deviation of each satellite product and of in situ data.

For the comparison with the Level 3 products, in order to be able to compare the statistics between the different SSS products, we regridded all SSS products onto a 0.25 degree resolution grid and 1 degree resolution grid. The gridding at 1 degree resolution is achieved by averaging the SSS values in 1x1degree cells, while the gridding at 0.25 degree resolution is achieved by oversampling the products using a bi-linear interpolation. TSG data are averaged either at 0.25 degree or 1 degree resolution.


Statistics have been estimated during the overlapping period between SMOS and Aquarius (from 09-2011 to 12-2012). That period includes 21 ship SSS transects. We have checked that the obtained statistics for SMOS and ISAS were not significantly modified by using TSG data from 07-2011 to 12-2012. For all monthly satellite products, a combination of ascending and descending orbits was used. In these comparisons we did not take into account uncertainty on TSG data nor on ISAS fields. The uncertainty on TSG data is usually around 0.01-0.02 and is much less compared to the satellite uncertainty and thus are not taken into account in the comparisons. ISAS maps include an estimation of the error, expressed as percentage of a priori variance. But we did not find any relationship between the error in ISAS and the statistics of our colocations.

Case study 5: Equatorial Pacific Further details and references are included in:

Yin, X., J. Boutin, G. Reverdin, T. Lee, S. Arnault, N. Martin, SMOS Sea Surface Salinity signals of tropical instability waves, Journal Geophys. Res.-Oceans, 2014, in revision.

SMOS level 3 SSS maps (combined ascending and descending orbits) averaged with 10-day, 100 x 100 km2 running windows and sampled daily over a 0.25 x 0.25° grid have been generated at Laboratoire d’Océanographie et du Climat: Expérimentation et Approches Numériques (LOCEAN) (http://catds.ifremer.fr/Products/Available-products-from-CEC-OS/Locean-v2013) [Boutin et al., 2013; Yin et al., 2012a]. The SMOS SSS before June 2010 are not used due to the variable configurations for testing the functionalities of the instrument and low-level procedures for data acquisition and handling during In-Orbit Commissioning Phase [Corbella et al., 2011].

We also use daily SST from the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA) produced on an operational basis at the UK Met Office using optimal interpolation on a global 0.054 degree grid [Donlon et al., 2011] (http://ghrsst-pp.metoffice.com/pages/latest_analysis/ostia.html). The SST maps were resampled over the same 0.25 x 0.25° grid as the SSS maps.

Near 5°N, the SMOS SSS vortices associated with TIWs are well observed together with clockwise rotating patterns of currents taken from Ocean Surface Current Analyses – Real time [Bonjean and Lagerloef, 2002] (OSCAR, http://www.oscar.noaa.gov/) (Figure 1a). The combined action by a series of tropical instability vortices (TIVs) is likely togenerate the cusp-shaped wave patterns associated with TIWs in the SMOS SSS and the OSTIA SST (Figure 1b). The SSS cusps are particularly noticeable in the northern hemisphere, with the northward higher SSS cusps coinciding with that of the lower SST.


Figure 1. 10-day average SMOS SSS (color shading), (a) 5-day average OSCAR currents (vectors) and (b) daily OSTIA SST (isolines) centered on September 24, 2010. The white squares on 1b correspond to the TAO moorings used for comparisons of SSS and SST.

Due to the periods generally associated with TIWs (20-40 days), SSS and SST signals with period longer than 50 days, in particular the seasonal variability, are removed by 50-day high-pass (50d-HP) filtering. Then SSS and SST signals associated with TIWs are isolated using 28-40-day band-pass filtering and 13-22-day band-pass filtering, corresponding to 33- and 17-day periods, respectively. The corresponding signals will be so-called the 50d-HP, 33- and 17-day signals in the following. The results are robust and remain unchanged when we alter the period bands slightly (45- or 55-day high-pass, and 25-40-day or 10-25 day band-pass).

Data from the Tropical Atmosphere Ocean (TAO) mooring project are also used in this analysis. They consist in SSS and SST daily data at 1m depth from June to December 2010. They are filtered using the same high-pass filtering and band-pass filtering as previously described. Filtered TAO data are then compared with SMOS SSS and OSTIA SST signals associated with TIWs. Six moorings with time series longer than 100 days from June to December 2010 and with peak amplitude of 33-day SSS signals larger than 0.1 are selected for comparisons (Table 1). Only high quality flagged (flag = 1) TAO SSS and SST [Henocq et al., 2010] are kept. TAO SSS at [140°W, 2°N] during mid-September and December 2010 are not used due to strong anomalies with bad quality controls (Paul Freitag, personal comm.).

The official release of Aquarius Combined Active Passive (CAP) algorithm level 3 sea surface salinity standard mapped image 7-day data v2.0 [Yueh et al., 2013] (http://podaac.jpl.nasa.gov/dataset/AQUARIUS_L3_SSS_CAP_7DAY_V2) are used for the long-term inter-comparison between SMOS and Aquarius SSS during September 2011 and June 2013.


4 Tools and systems to be implemented

This section of the TS gives a technical description of the selected tools and system(s) (including models, databases, visualisation tools, related data sources, processing steps and output data) required to perform the SOS project. It is expected that individual researchers will use other “common” tools at their laboratory (e.g. Python, Matlab, IDL etc) and these are considered “normal business” and are not detailed in the TS.

4.1 Tools and systems by partner This section presents the software, hardware and processing activities to be carried out in order to address the case study requirements (CSREQ) given in the SAP conclusions chapter.

4.1.1 LOCEAN systems LOCEAN computing system is composed of a network of Linux workstation, 4 highperformance servers dedicated to scientific computation. IPSL (Institut Pierre Simon Laplace)provides us a cluster with 256 processors linked with a large filesystem (~800To) and a datacenter where ESA SMOS-Ocean Salinity level 1 and 2 and CATDS products are available.LOCEAN is connected to the Internet via a 10 gigabit line to RENATER. All processings performed at LOCEAN are done using python scripts under LOCEAN workstations network under LINUX.

Task Processing software

Hardware CS adressed CSREQ addressed

L2 & L3 SMOS processing, including bias mitigation

Python scripts

LOCEAN workstations network under LINUX

CS3-10, CS4-30, CS5-10, CS5-20,CS5-30

3-20, 4-10, 4-50, 4-60

SMOS-AQUARIUS comparisons

Python scripts


CS4-30 4-75

Colocations between SMOS products, Rain products and in situ products

Python scripts


CS3-20, CS4-30, CS5-20,CS5-30

3-10, 3-15,3-30, 3-35, 3-40,4-10, 4-55, 4-80, 5-20, 5-30

Data visualization

Python scripts


CS3-10, CS3-20, CS4-30, CS5-10, CS5-20,CS5-30

3-20, 4-10, 4-55, 4-60, 4-65, 4-70, 4-80, 5-10, 5-20, 5-25, 5-40


4.1.2 Ifremer systems



All where ifremer is involved

C+ compiled Matlab

Cloud computing

CS1 CSREQ-1-25, 1-30, 1-35, 1-45, 1-70, 1-75, 1-80, 1-85, 1-95

4.1.3 NOC systems NOC has a state of the art computing system comprising a network of SUN, Linux and Apple workstations linked via a high speed LAN to 45 TBytes of on-line disc space and a 400 TByte tape store. In addition to workstations, there is a 96-processors Bull Linux cluster and supercomputer used to run global ocean circulation and climate models.



L2 data collection routines

Shell scripts Linux workstations

CS2, CS4 CSREQ-2-30, CSREQ-4-10, CSREQ-4-30, CSREQ-4-45

SMOS/ Aquarius L3 gridding

Matlab Linux workstations

CS2, CS4 CSREQ-2-55, CSREQ-4-40

4.1.4 MetOffice systems The Met Office computing system is composed of a network of Linux workstations, various Linux compute servers, and an IBM high performance computing (HPC) facility (supercomputer). An archiving system (MASS) is used to store the vast amounts of data produced and contains the FOAM fields which are to be used within this project.

Processing of data is generally done using either Fortran, IDL or Python on the Met Office Linux workstations. NetCDF operators are used for some data extraction and processing tasks. Visualisation is now being done using Python (although some plots are produced with IDL).



Extraction of relevant fields for use in the project

Unix scripts and NetCDF

operators

Linux workstations

CS2-10,CS2-20, CS2-30

2-10,2-15,2-20,2-35,2-50,2-55,2-80,4-15

Collocate various data sources including

Fortran code Linux workstations

CS2-10 2-10,2-50,2-55


FOAM/Argo/SMOS and FOAM/Argo/Aquarius

Calculating positions of Gulf Stream and Agulhas fronts from various data-sets

Fortran code Linux workstations

CS2-20,CS2-30

2-15,2-20,2-70,2-75,2-80

Visualisation Python Linux workstations

CS2-10,CS2-20, CS2-30

2-15,2-20,2-25,2-30,2-35,2-40,2-45,2-50,2-55,2-60,2-65,2-70,2-75,2-80

4.2 Tools Design Justification This section presents a brief justification for the tools and systems selected for use in the project.

Tools that will be used at LOCEAN have already been widely used for same kind of work (see publications by Boutin et al 2012, 2013). Tools foreseen to be used at IFREMER have already been used for various similar technical tasks (e.g. Reul et al. 2009; Grodsky et al. 2012)

The collocation tools used at the Met Office have already been used for various similar tasks including for the operational interpolation of model fields to observation locations, and match-up code used in FOAM for SST bias correction.

NOC L3 gridding systems have been used to provide products for similar work in the past.

4.3 Risks and proposed solutions Tools and systems risk Proposed solution

No identified risk at LOCEAN

No identified risk at IFREMER

No identified risk at Met Office

No identified risk at NOC

24 June 2014 63 System development requirements

5 System development requirements

This section presents an analysis of specific development activities required to implement the case study experiments as specified in the SAP.

5.1 Development requirements by Case Study Experiment

This section summarises the technical development to be carried out during the project to complete each Case Study Experiment.

Experiment Development required Responsible CSREQ addressed

CS1-10 None: all tools already available NR 1-25,1-30,1-35

CS1-20 None: all tools already available NR 1-45

CS1-30 None: all tools already available NR 1-70, 1-75, 1-80, 1-85, 1-95

CS2-10 Existing interpolation and match-up code already in use at the Met Office will be adapted.

MM 2-10

CS2-20 Existing code to calculate gradients used in OSTIA will be adapted.

MM 2-15

CS2-30 Existing code to calculate gradients used in OSTIA will be adapted.

MM 2-20

CS3-10 No- Adaptation of tools existing at LOCEAN

JB 3-20, 3-25

CS3-20 Handling of STS floats data JB 3-15, 3-30, 3-35

CS3-30 Existing interpolation and match-up code already in use at the Met Office will be adapted.

MM 3-30

CS4-10 CS4-20

Requirement to develop tools for processing OSTIA SST data

CB 4-30, 4-40, 4-45

CS4-10 CS4-20

Requirement to develop tools for processingTRMM Rainfall

CB 4-20, 4-45


JB 4-10, 4-50, 4-60, 4-65, 4-70, 4-75, 4-80


JB 5-20


JB 5-25, 5-30

CS5-30 Handling of MAPCO2 data JB 5-35, 5-40

24 June 2014 64 System development requirements

5.2 Development Design Justification Most technical tools have already been used in previous similar studies (see section 3.2) and found to be efficient. The adaptations required to address the SMOS SOS project are considered normal scientific business and are not described.

5.3 Risks and proposed solutions

System development risk Proposed solution

No identified risk at LOCEAN

No identified risk at IFREMER

No identified risk at Met Office

No identified risk at NOC

24 June 2014 65 Interface control for the project

6 Interface control for the project

This section presents the data interfaces and experimental data set required for the project.

6.1 Data interface matrix This section summarises the data interfaces for each Case Study Experiment. Origin of data reported in the Table is abbreviated. Please refer to section 2 for a full description of the origin of each data set.

EO-01 SMOS L3 : CATDS CEC

EO-02 SMOS L3 CATDS-CEC-LOCEAN

EO-03 SMOS L2

EO-04 AQUARIUS -NASA

EO-05 AQUARIUS L3-NOC

EO-06 AMSR-E

EO-07 MODIS

EO-08 GLOBCOLOR

EO_09 OSCAR

EO-10 SSMI Rain

EO-11 TRMM3B42

EO-12 AVHRR (METOP)

EO-13 OSTIA

EO-14 ASCAT

CS1-10

Internal at IFREMER;

ftp from CATDS-CEC-IFREMER






CS1-20







CS1-30







CS1-40







CS2-10

ftp from CATDS-CEC to Metoffice & NOC

ftp from CATDS-CEC-LOCEAN to Metoffice &NOC

ftp from Brookman ftp site

ftp from PODAAC to Metoffice & NOC

Internal at NOC

- - - - ftp from ftp://ftp.ssmi.com/ssmi to Metoffice

-


CS2-20


ftp from CATDS-CEC-LOCEAN to Metoffice & NOC



Internal at NOC

Internal at Metoffice

ftp from PODAAC to Metoffice

- Download from http://www.oscar.noaa.gov to Metoffice

- - Internal to Met Office

CS2-30


ftp from CATDS-CEC-LOCEAN to Metoffice & NOC



Internal at NOC

Internal at Metoffice

ftp from PODAAC to Metoffice

- Download from http://www.oscar.noaa.gov to Metoffice

- - Internal to Met Office

CS3-10

Internal at LOCEAN;


ftp from ssmi.com

ftp from giovanni

CS3-20


ftp from ssmi.com

ftp from giovanni

CS3-30


CS4-10

ftp from CATDS-CEC to NOC


ftp from giovanni

CS4-20

Internal at NOC

CS4-30

ftp from CATDS-CEC to LOCEAN

ftp from CATDS-CEC to NOC


Delivery by ftp from NASA web site

CS5-10

ftp from CATDS-CEC to LOC

Internal at LOCEAN;


EAN

CS5-20


Internal at LOCEAN;


CS5-30


Internal at LOCEAN;


IS-01 ARGO

IS-02 ARGO-STS

IS-03 ORE-SSS

IS-04 Drifters

IS-05 ISAS

IS-06 STRASSE

IS-07 TAO/TRITON

IS-08 MAPCO2

IS-09 NODC

IS-10 ORE-HYBAM

NM-01 FOAM-Global

NM-02 FOAM-N. ATl

CS1-10 Ifremer

CS1-20 Ifremer ftp from CATDS-CEC-IFREMER

CS1-30 ftp from CATDS-CEC-IFREMER

CS1-40 ftp from CATDS-CEC-IFREMER

CS2-10 Use EN3 data Internal at Metoffice

- ftp from ORE-SSS

Internal Metoffice

- - - - Internal Metoffice

Internal Metoffice

CS2-20 - - ftp from ORE-SSS

Internal Metoffice


Internal Metoffice

CS2-30 - - ftp from ORE-SSS

Internal Metoffice


-

CS3-10 ftp from coriolis

Internal LOCEAN

Internal Metoffice

ftp from ssmi.com

CS3-20 Courtesy of S. Reiser

Internal at LOCEAN

ftp from ssmi.com

CS3-30

CS4-10 Internal Metoffice

Internal Metoffice


CS4-20

CS4-30 ftp from ORE-SSS

Internal LOCEAN

Internal LOCEAN

CS5-10 Internal LOCEAN

CS5-20 ftp noaa.tao

CS5-30 ftp cdiac

6.2 Experimental Scientific Dataset This section presents the satellite and in situ data products that will constitute the experimental dataset for each Case Study.

6.2.1 CS1 Scientific Dataset Label Description Location Time period Format

EO-01 SMOS L3 : CATDS CEC 70°W-30°W;10°S-30° 2010-2012 NetCDF

EO-06 AMSR-E 70°W-30°W;10°S-30°N

2010-2011 See DARD table

EO-07 MODIS 70°W-30°W;10°S-30° 2010-2012 See DARD table

EO-08 GLOBCOLOR 70°W-30°W;10°S-30° 2010-2012 See DARD table

EO-09 OSCAR Currents 70°W-30°W;10°S-30° 2010-2012 See DARD table

EO-14 ASCAT Wind Speed 70°W-30°W;10°S-30° 2010-2012 See DARD table


EO-01 SMOS L3 : CATDS CEC Gulf Stream: 90W – 30W, 20N – 55N.

Agulhas: 0E – 50E, 45S – 20S.

1st Dec 2010 – 30th Nov 2012.

NetCDF

EO-05 AQUARIUS L3-NOC Gulf Stream: 90W – 30W, 20N – 55N.

Agulhas: 0E – 50E, 45S – 20S.

1st Dec 2010 – 30th Nov 2012.

See DARD table

EO-09 OSCAR Currents Gulf Stream: 90W – 30W, 20N – 55N.

Agulhas: 0E – 50E, 45S – 20S.

1st Dec 2010 – 30th Nov 2012.

See DARD table

IS-01 ARGO Gulf Stream: 90W – 30W, 20N – 55N.

1st Dec 2010 – 30th Nov 2012.

See DARD table


Agulhas: 0E – 50E, 45S – 20S.

NM-01 FOAM-Global Gulf Stream: 90W – 30W, 20N – 55N.

Agulhas: 0E – 50E, 45S – 20S.

1st Dec 2010 – 30th Nov 2012.

See DARD table

NM-02 FOAM-N.Atl. Gulf Stream: 90W – 30W, 20N – 55N.

Agulhas: 0E – 50E, 45S – 20S.

1st Dec 2010 – 30th Nov 2011.

See DARD table


EO-02

SMOS L3 CATDS-CEC-LOCEAN

30N-30S 2010-2011 NetCDF

EO-03 SMOS L2 30N-30S 2010-2011 See DARD table

EO-10 SSM/I Rain 30N-30S 2010-2011 See DARD table

EO-11 TRMM 3B42 30N-30S 2010-2011 See DARD table

IS-01 ARGO 30N-30S 2010-2011 See DARD table

IS-02 ARGO-STS 30N-30S 2010-2011 See DARD table

IS-04 Drifters 30N-30S 2010-2011 See DARD table

IS-05 ISAS 30N-30S 2010-2011 See DARD table

NM-01 FOAM-Global 30N-30S 2010-2011 See DARD table


EO-02


15N-35N ; 50W-20W 2010-2012 NetCDF

EO-03 SMOS L2 15N-35N ; 50W-20W 2010-2012 See DARD table

EO-04 AQUARIUS L3-NASA 15N-35N ; 50W-20W Sep2011-2012 See DARD table

EO-05 AQUARIUS L3-NOC 15N-35N ; 50W-20W Sep2011-2012 Matlab MAT

EO-11 TRMM 3B42 15N-35N ; 50W-20W 2010-2012 See DARD table

IS-05 ISAS 15N-35N ; 50W-20W 2010-2012 See DARD table

IS-06 STRASSE 15N-35N ; 50W-20W Aug-Sep 2012 See DARD table

NM-01 FOAM-Global 15N-35N ; 50W-20W 1st Dec 2010 – 30th Nov 2012.

See DARD table



EO-01

SMOS L3 : CATDS CEC 30N-30S 2010-2011 NetCDF

EO-02


30N-30S 2010-2011 NetCDF

EO-03 SMOS L2 30N-30S 2010-2011 See DARD table

IS-05 ISAS 30N-30S 2010-2011 See DARD table

IS-07 TAO/TRITON 10N-10S 2010-2011 See DARD table

IS-08 MAP CO2 10N-10S (see DARD) 2010-2011 See DARD table

6.3 Interface Design Justification This section presents a brief justification for the interfaces planned in the project.

All interfaces are prescribed due to location of datasets and requirements for scientific work. As such there is no specific design of interfaces.

6.4 Risks and proposed solutions

Interface risk Proposed solution

None identified risk at LOCEAN

None identified risk at IFREMER

None identified risk at Met Office

None identified risk at NOC

24 June 2014 71 Risk analysis and mitigation

7 Risk analysis and mitigation

This section presents a summary of risks given in the risk tables in above sections. In general few risks have been identified except for some relating to availability of data:

Data risk Proposed solution

IS-02: These data are not yet in Coriolis data base but should be put before end 2013 (T. Carval, Coriolis, pers. Comm.)

We already got an agreement from Steve Reiser (PI of ARGO_STS) to access the original files.

IS-08: Only real time data is visible on web site after 2009

In case no verified data is available when we’ll start our study (in 2014) we’ll take contact with NOAA responsible and eventually conduct rough validation based on earlier measurements in the area. An alternative could be to focus into area covered by the SOCAT version 2 pCO2 data base recently released (http://www.socat.info/about.html).

24 June 2014 72 Summary and conclusions

8 Summary and conclusions

The above sections present the technical specifications foreseen to carry out the Case Study experiments in the SMOS+SOS project. Most data and tools already exist and are available in the laboratories of partners involved in the project. They could however evolve further depending on the scientific issues found during the course of the project. There are no major technical risks associated with the project at the time of writing.

24 June 2014 73 References

9 References

Beal, L. M., W. P. M. De Ruijter, A. Biastoch, R. Zahn, and S. W. I. W. Grp (2011), On the role of the Agulhas system in ocean circulation and climate, Nature, 472(7344), 429-436.

Blockley, E., R. Barciela, R. Furner, C. Guiavarc'h, A. Hines, D. Lea, R. Mahdon, M. Martin, and D. Storkey (2008), Regional FOAM Configurations, Final GODAE Symposium 2008 “The revolution in global ocean forecasting: GODAE: 10 years of achievement".

Blockley, E. W., M. J. Martin, and P. Hyder (2012), Validation of FOAM near-surface ocean current forecasts using Lagrangian drifting buoys, Ocean Science, 8(4), 551-565.

Boutin, J., and N. Martin (2006), ARGO upper salinity measurements: Perspectives for L-band radiometers calibration and retrieved sea surface salinity validation, IEEE Geoscience and Remote Sensing Letters, 3(2), 202-206.

Boutin, J., L. Merlivat, C. Henocq, N. Martin, and J. B. Sallee (2008), Air-sea CO(2) flux variability in frontal regions of the Southern Ocean from CARbon Interface OCean Atmosphere drifters, Limnology and Oceanography, 53(5), 2062-2079.

Boutin, J., N. Martin, G. Reverdin, X. Yin, and F. Gaillard (2012), Sea surface freshening inferred from SMOS and ARGO salinity: impact of rain, Ocean Sci. Discuss., 9(5), 3331-3357.

Boutin, J., N. Martin, X. Yin, J. Font, N. Reul, and P. Spurgeon (2012), First Assessment of SMOS Data Over Open Ocean: Part II-Sea Surface Salinity, Ieee Transactions on Geoscience and Remote Sensing, 50(5), 1662-1675.

Boutin, J., et al. (1999), Satellite sea surface temperature: a powerful tool for interpreting in situ pCO2 measurements in the equatorial Pacific Ocean, Tellus B, 51(2), 490-508.

Bryden, H. L., H. R. Longworth, and S. A. Cunningham (2005), Slowing of the Atlantic meridional overturning circulation at 25 degrees N, Nature, 438(7068), 655-657.

CLIVAR, U. S. (2007), Report of the US CLIVAR salinity science working groupRep., 46 pp, U.S. CLIVAR, Washington DC.

Cravatte, S., T. Delcroix, D. Zhang, M. J. McPhaden, and J. Leloup (2009), Observed freshening and warming of the western Pacific warm pool, Climate Dyn., 33, 565-589.

Donlon, C. J., M. Martin, J. Stark, J. Roberts-Jones, E. Fiedler, and W. Wimmer (2012), The Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA) system, Remote Sensing of Environment, 116, 140-158.

Drecourt, J.-P., K. Haines, and M. Martin (2006), Influence of systematic error correction on the temporal behavior of an ocean model, Journal of Geophysical Research-Oceans, 111(C11).

Durack, P. J., S. E. Wijffels, and R. J. Matear (2012), Ocean Salinities Reveal Strong Global Water Cycle Intensification During 1950 to 2000, Science, 336(6080), 455-458.


Eldin, G., T. Delcroix, and M. Rodier (2004), The frontal area at the eastern edge of the western equatorial Pacific warm pool in April 2001, J. Geophys. Res., 109(C7), C07006.

Font, J., G. S. E. Lagerloef, D. M. Le Vine, A. Camps, and O. Z. Zanife (2004), The determination of surface salinity with the European SMOS space mission, IEEE Transactions on Geoscience and Remote Sensing, 42(10), 2196-2205.

GCOS (2011), 'GCOS Ocean Surface ECV Sea Surface Salinity (SSS), edited.

Gentemann, C. L., C. J. Donlon, A. Stuart-Menteth, and F. J. Wentz (2003), Diurnal signals in satellite sea surface temperature measurements, Geophysical Research Letters, 30(3).

Gordon, A. L., and C. F. Giulivi (2008), Sea surface salinity trends over fifty years within the subtropical north Atlantic, Oceanography, 21(1), 20–29.

Grodsky, S. A., N. Reul, G. Lagerloef, G. Reverdin, J. A. Carton, B. Chapron, Y. Quilfen, V. N. Kudryavtsev, and H.-Y. Kao (2012), Haline hurricane wakein the Amazon/Orinoco plume: AQUARIUS/SACD and SMOSobservations, Geophys. Res. Lett., 39, L20603, doi:10.1029/2012GL053335.

Gould, J., et al. (2004), Argo Profiling Floats Bring New Era of In Situ Ocean Observations, EOS Trans, 85(19)(185).

Henocq, C., J. Boutin, G. Reverdin, F. Petitcolin, S. Arnault, P. Lattes, 2010: Vertical Variability of Near-Surface Salinity in the Tropics: Consequences for L-Band Radiometer Calibration and Validation. Journal of Atmospheric and Oceanic Technology, 27, 192-209.

Hyder, P., D. Storkey, E. Blockley, C. Guiavarc'h, J. Siddom, M. Martin, and D. Lea (2012), Assessing equatorial surface currents in the FOAM Global and Indian Ocean models against observations from the global tropical moored buoy array, J. Oper. Oceanogr., 5(2), 25-39.

IPCC (2007), IPCC Fourth Assessment Report: Climate Change 2007 (AR4)Rep., Zurich.

Kwon, Y.-O., M. A. Alexander, N. A. Bond, C. Frankignoul, H. Nakamura, B. Qiu, and L. Thompson (2010), Role of the Gulf Stream and Kuroshio-Oyashio Systems in Large-Scale Atmosphere-Ocean Interaction: A Review, Journal of Climate, 23(12), 3249-3281.

Lagerloef, G., R. Schmitt, J. Schanze, and H.-Y. Kao (2010), The Ocean and the Global Water Cycle, Oceanography, 23(4), 82-93.

Lagerloef, G., et al. (2010), Resolving the Global Surface Salinity Field and Variations by Blending Satellite and In Situ Observations, in Oceanobs'09: Sustained Ocean Observations and Information for Society edited by J. Hall, Harrison, D.E. & Stammer, D., ESA Publication WPP-306, Venise, Italy, 21-25 September 2009.

Le Borgne, R., R. T. Barber, T. Delcroix, H. Y. Inoue, D. J. Mackey, and M. Rodier (2002), Pacific warm pool and divergence: temporal and zonal variations on the equator and their effects on the biological pump, Deep Sea Research Part II: Topical Studies in Oceanography, 49(13–14), 2471-2512.

Lea, D. J., J. P. Drecourt, K. Haines, and M. J. Martin (2008), Ocean altimeter assimilation with observational- and model-bias correction, Quarterly Journal of the Royal Meteorological Society, 134(636), 1761-1774.


Lee, T., G. Lagerloef, M. M. Gierach, H.-Y. Kao, S. Yueh, and K. Dohan (2012), Aquarius reveals salinity structure of tropical instability waves, Geophysical Research Letters, 39.

Maes, C., J. Sudre, and V. Garçon (2010), Detection of the Eastern Edge of the Equatorial Pacific Warm Pool Using Satellite-Based Ocean Color Observations, Scientific Online Letters on the Atmosphere, 6, 129-132.

Martin, A. J., A. Hines, and M. J. Bell (2007), Data assimilation in the FOAM operational short-range ocean forecasting system: A description of the scheme and its impact, Quarterly Journal of the Royal Meteorological Society, 133(625), 981-995.

Mirouze, I., and A. T. Weaver (2010), Representation of correlation functions in variational assimilation using an implicit diffusion operator, Quarterly Journal of the Royal Meteorological Society, 136(651), 1421-1443.

Mogensen, K. S., M. A. Balmaseda, A. Weaver, M. J. Martin, and A. Vidard (2009), NEMOVAR: A variational data assimilation system for the NEMO ocean model, ECMWF newsletter, Summer 2009.

O'Dea, E. J., et al. (2012), An operational ocean forecast system incorporating NEMO and SST data assimilation for the tidally driven European North-West shelf, J. Oper. Oceanogr., 5(1), 3-17.

Oke, P. R., G. B. Brassington, J. Cummings, M. Martin, and F. Hernandez (2012), GODAE inter-comparisons in the Tasman and Coral Seas, J. Oper. Oceanogr., 5(2), 11-24.

Physical Oceanography Distributed Active Archive Center (2012), Aquarius User Guide v3.7Rep., Jet Propulsion Laboratory, Pasadena, California, USA

Picaut, J., M. Ioualalen, T. Delcroix, F. Masia, R. Murtugudde, and J. Vialard (2001), The oceanic zone of convergence on the eastern edge of the Pacific warm pool: A synthesis of results and implications for El Niño-Southern Oscillation and biogeochemical phenomena, J. Geophys. Res., 106(C2), 2363-2386.

Quartly, G. D., and M. A. Srokosz (2002), SST observations of the Agulhas and East Madagascar retroflections by the TRMM Microwave Imager, Journal of Physical Oceanography, 32(5), 1585-1592.

Read, J. F., and R. T. Pollard (1993), STRUCTURE AND TRANSPORT OF THE ANTARCTIC CIRCUMPOLAR CURRENT AND AGULHAS RETURN CURRENT AT 40-DEGREES-E, Journal of Geophysical Research-Oceans, 98(C7), 12281-12295.

Read, J. F., M. I. Lucas, S. E. Holley, and R. T. Pollard (2000), Phytoplankton, nutrients and hydrography in the frontal zone between the Southwest Indian Subtropical gyre and the Southern Ocean, Deep-Sea Research Part I-Oceanographic Research Papers, 47(12), 2341-2368.

Reul, N., S. Saux-Picart, B. Chapron, D. Vandemark, J. Tournadre, and J. Salisbury (2009), Demonstration of oceansurface salinity microwave measurements from space usingAMSR-E data over the Amazon plume, Geophys. Res. Lett., 36,L13607, doi:10.1029/2009GL038860.

Reul, N., and J. Tenerelli (2010), Water mission reveals insight into Amazon plume, edited, ESA.


Reul, N., et al. (2005), Synergetic aspects and auxiliary data concepts for sea surface salinity measurements from space, Final Report to ESA, Contract No. 18176/04/NL/CB,Rep., 613 pp, IFREMER.

Reverdin, G., S. Morisset, J. Boutin, and N. Martin (2012), Rain-induced variability of near sea-surface T and S from drifter data, J. Geophys. Res., 117(C2), C02032.

Riser, S., and J. Anderson (2012), STS Argo Measurements, 7th Aquarius/SAC-D Science Meeting Preliminary.

Roemmich, D., and J. Gilson The 2004-2008 mean and annual cycle of temperature, salinity, and steric height in the global ocean from the Argo Program, Progress in Oceanography, In Press, Corrected Proof.

Soloviev, A., and R. Lukas (1996), Observation of spatial variability of diurnal thermocline and rain-formed halocline in the western Pacific warm pool, Journal of Physical Oceanography, 26(11), 2529-2538.

Stark, J. D., C. J. Donlon, M. J. Martin, and M. E. McCulloch (2007), OSTIA : An operational, high resolution, real time, global sea surface temperature analysis system., paper presented at Oceans '07 IEEE Marine challenges: coastline to deep sea, IEEE, Aberdeen.

Stark, J. D., C. Donlon, A. O'Carroll, and G. Corlett (2008), Determination of AATSR biases using the OSTIA SST analysis system and a matchup database, Journal of Atmospheric and Oceanic Technology, 25(7), 1208-1217.

Storkey, D., E. W. Blockley, R. Furner, C. Guiavarc'h, D. Lea, M. J. Martin, R. M. Barciela, A. Hines, P. Hyder, and J. R. Siddorn (2010), Forecasting the ocean state using NEMO:The new FOAM system, J. Oper. Oceanogr., 3(1), 3-15.

Suess, M., P. Matos, A. Gutierrez, M. Zundo, and M. Martin-Neira (2004), Processing of SMOS level 1C data onto a Discrete Global Grid, paper presented at IGARSS 2004: IEEE International Geoscience and Remote Sensing Symposium Proceedings, Vols 1-7 - Science for Society: Exploring and Managing a Changing Planet, IEEE, New York.

Tenerelli, J. E., N. Reul, A. A. Mouche, and B. Chapron (2008), Earth-viewing L-band radiometer sensing of sea surface scattered celestial sky radiation - Part I: General characteristics, IEEE Transactions on Geoscience and Remote Sensing, 46(3), 659-674.

Terray, L., L. Corre, S. Cravatte, T. Delcroix, G. Reverdin, and A. Ribes (2011), Near-Surface Salinity as Nature’s Rain Gauge to Detect Human Influence on the Tropical Water Cycle, Journal of Climate, 25(3), 958-977.

Ulaby, F. T., R. K. Moore, and A. K. Fung (1986), Microwave remote sensing: active and passive, 1117 pp., Artech House, Inc, Dedham, MA, USA.

Vialard, J., et al. (2009), Cirene: Air—Sea Interactions in the Seychelles—Chagos Thermocline Ridge Region, Bulletin of the American Meteorological Society, 90(1), 45-61.

Weaver, A. T., C. Deltel, E. Machu, S. Ricci, and N. Daget (2005), A multivariate balance operator for variational ocean data assimilation, Quarterly Journal of the Royal Meteorological Society, 131(613), 3605-3625.


Yu, L. S. (2010), On Sea Surface Salinity Skin Effect Induced by Evaporation and Implications for Remote Sensing of Ocean Salinity, Journal of Physical Oceanography, 40(1), 85-102.

Zhang, Y., and X. Zhang (2012), Ocean haline skin layer and turbulent surface convections, J. Geophys. Res., 117(C4), C04017.

24 June 2014 78 Appendices

10 Appendices

There are no appendices.