ESA SSS CCI v5.5¶

ESA Sea Surface Salinity Climate Change Initiative (Sea_Surface_Salinity_cci) — Weekly and monthly sea surface salinity products from L-band, v5.5 (merged SMOS, Aquarius and SMAP Level 4 sea surface salinity climate record, distributed by CEDA)

Available

At a glance¶

Note on period: The CEDA records give 2010-01-01 to 2023-12-30. The global monthly files run from 20100115 to 20231215, the last global weekly file is 2023-12-30, and the polar weekly files start on 2010-01-02 (Southern Hemisphere) and 2010-01-09 (Northern Hemisphere). Input missions are SMOS (2010 onward), Aquarius (2011-08 to 2015-06) and SMAP (2015-04 onward), according to the v4.41 PUG. January to May 2010 is built from SMOS commissioning-phase data, and the PUG notes that products are generally better after 2015, once SMAP is included.

Overview¶

A Level 4 climate record from the ESA Climate Change Initiative (CCI) that cross-calibrates and merges sea surface salinity from three L-band radiometer missions: SMOS (ESA), Aquarius and SMAP (NASA). Biases between sensors are removed and each pixel's long-term mean is anchored to the long-term mean of the in situ ISAS salinity analysis, giving 50 km resolution weekly (7-day running mean, daily) and monthly (every 15 days) salinity fields on a global 0.25° grid and on 25 km polar EASE-2 grids. Every value carries a random-error estimate and quality flags. Version 5.5 reprocesses 2010–2023 with a new SMOS dielectric constant model, SMAP RSS v5.3 input and global RFI mitigation.

What changes with the successor¶

SSS CCI v4.41 → SSS CCI v5.5¶

Transition: 2025-02

Extends the record by one year to 2023 and brings in reprocessed SMOS data (new dielectric model), a newer SMAP input version and global RFI mitigation. The 2010 underestimation and the April–August 2015 salty error in the North Atlantic found in v4.41 were fixed, and the Northern Hemisphere seasonal bias was reduced.

Sources: ESA CCI — Sea Surface Salinity Record Extended (v5.5) · SSS CCI PVIR v5.1 (2025-02-25) · SSS CCI PUG v4.1 (known issues in v4.41)

Available variables¶

Variable names follow the CEDA v5.5 records and the ncdump in the PUG. Weekly and monthly files contain the same variables. The salinity units attribute is "0.001" (practical salinity values, typically about 30–38), and the quality flags use 0 for good and 1 for bad.

instantaneous mean accumulated maximum minimum static — instantaneous: value at the time stamp · mean/accumulated/maximum/minimum: over the indicated interval · static: does not change with time

Weekly (7-day running mean, daily)¶

Monthly (centred on the 1st and 15th)¶

The full variable list is in the official documentation.

How to get the data¶

• CEDA Archive — SSS CCI v5.5 collection (HTTPS, FTP, OPeNDAP) — Direct HTTPS download / anonymous FTP (anon-ftp.ceda.ac.uk) / OPeNDAP (netCDF) No login needed
  • No login is needed. Files are under https://data.ceda.ac.uk/neodc/esacci/sea_surface_salinity/data/v05.5/ in GLOBALv5.5 (0.25°) and NHv5.5 and SHv5.5 (25 km EASE-2), each split into 7days (weekly) and 30days (monthly) folders by year. Example: GLOBALv5.5/30days/2023/ESACCI-SEASURFACESALINITY-L4-SSS-GLOBAL-MERGED_OI_Monthly_CENTRED_15Day_0.25deg-20230115-fv5.5.nc
• CCI THREDDS server (OPeNDAP and catalogue) — OPeNDAP / HTTP (folders v05.5, v04.41, v03.21, v02.31, v01.8) No login needed
  • Lets you read just the region you need with xarray and similar tools without downloading whole files. Replace catalog with dodsC in the path to get the OPeNDAP address (see the example below).

Example code¶

Reads one monthly file over OPeNDAP from the CCI THREDDS server without logging in, and clips the seas around Korea. Requiring all three flags to be 0 is one conservative choice.

# pip install xarray netCDF4
import xarray as xr

base = ("https://data.cci.ceda.ac.uk/thredds/dodsC/esacci/sea_surface_salinity/"
        "data/v05.5/GLOBALv5.5/30days/2023/")
fn = "ESACCI-SEASURFACESALINITY-L4-SSS-GLOBAL-MERGED_OI_Monthly_CENTRED_15Day_0.25deg-20230815-fv5.5.nc"
ds = xr.open_dataset(base + fn).sortby("lat")

good = (ds["sss_qc"] == 0) & (ds["lsc_qc"] == 0) & (ds["isc_qc"] == 0)
sss = ds["sss"].where(good)
sub = sss.sel(lat=slice(25, 45), lon=slice(115, 145))   # Yellow Sea, East China Sea, East Sea
print(float(sub.mean()), int(sub.count()))

Practical tips¶

• L-band satellite salinity describes the top centimetre or so. The PUG explains that it usually represents the upper few metres, but for a few hours after rain or in strongly stratified water such as river plumes it can be fresher than Argo or ship measurements taken metres below.
• Each pixel's long-term mean is adjusted to the long-term mean of ISAS, an Argo-based in situ salinity analysis. The time variability comes from the satellites, but the mean itself is tied to in situ data, so comparing it with Argo or ISAS mean fields is not a fully independent validation.
• The PVIR reports robust standard deviations above 0.4 pss against Argo in coastal and river-plume regions. In seas close to land and strongly influenced by Changjiang (Yangtze) outflow, such as the Yellow Sea and East China Sea, always check lsc_qc and be careful with quantitative analysis.
• Both the file attributes and the PVIR note a seasonally varying bias (under 0.05 pss peak to peak) in the Northern Hemisphere. Keep this size in mind when studying seasonal amplitudes at northern mid and high latitudes.
• For monthly means, use the files dated the 15th; averaging in the files dated the 1st mixes in neighbouring months. Weekly files also overlap as running means, so count degrees of freedom carefully.
• The PUG (v4.1) explains that the polar 25 km EASE-2 files are nearest-neighbour interpolations of the global product. The finer grid does not change the 50 km effective resolution.
• Do not join v5.5 with v4.41 or older versions in one time series. Version 5.5 fixed the 2010 underestimation and the April–August 2015 North Atlantic error linked to SMAP, and it changed the SMOS dielectric model and the SMAP input version.
• The PVIR reports larger scatter in cold high-latitude water below 4°C, such as the Barents Sea, possibly from dielectric model errors or sea-ice contamination. For polar seas, check isc_qc together with sss_random_error.

Notices and change history¶

References for your paper¶

Dataset citation¶

CEDA SSS CCI v5.5 (weekly and monthly collection) — DOI 10.5285/7294d93479654c139770f13fae4142d1

@misc{ssscci_v55,
  author    = {Boutin, J. and Vergely, J.-L. and Reul, N. and Catany, R. and Jouanno, J. and Martin, A. and Rouffi, F. and Bertino, L. and Bonjean, F. and Corato, G. and G{\'e}vaudan, M. and Guimbard, S. and Khvorostyanov, D. and Kolodziejczyk, N. and Matthews, M. and Olivier, L. and Raj, R. and R{\'e}my, E. and Reverdin, G. and Supply, A. and Thouvenin-Masson, C. and Vialard, J. and Sabia, R. and Mecklenburg, S.},
  title     = {{ESA} Sea Surface Salinity Climate Change Initiative ({Sea\_Surface\_Salinity\_cci}): Weekly and monthly sea surface salinity products from L-band, v5.5},
  year      = {2025},
  publisher = {NERC EDS Centre for Environmental Data Analysis},
  doi       = {10.5285/7294d93479654c139770f13fae4142d1}
}

Previous version v04.41 (2010–2022) — DOI 10.5285/f2ca631f29a24c47a7e98654ddf2c7d9

Reference papers¶

CCI salinity data description paper (recommended by CEDA and the PUG) — DOI 10.1029/2021JC017676

@article{boutin2021sss,
  author  = {Boutin, J. and Reul, N. and Koehler, J. and Martin, A. and Catany, R. and Guimbard, S. and Rouffi, F. and Vergely, J. L. and Arias, M. and Chakroun, M. and Corato, G. and Estella-Perez, V. and Hasson, A. and Josey, S. and Khvorostyanov, D. and Kolodziejczyk, N. and Mignot, J. and Olivier, L. and Reverdin, G. and Stammer, D. and Supply, A. and Thouvenin-Masson, C. and Turiel, A. and Vialard, J. and Cipollini, P. and Donlon, C. and Sabia, R. and Mecklenburg, S.},
  title   = {Satellite-Based Sea Surface Salinity Designed for Ocean and Climate Studies},
  journal = {Journal of Geophysical Research: Oceans},
  volume  = {126},
  number  = {11},
  pages   = {e2021JC017676},
  year    = {2021},
  doi     = {10.1029/2021JC017676}
}

Acknowledgements¶

The ESA CCI Salinity terms and conditions require presentations and publications to acknowledge the ESA Climate Change Initiative and the Sea Surface Salinity CCI project (and the individual data providers) and to cite the dataset DOIs. Users are also asked to send copies of reports and publications to the CCI programme and to consider offering co-authorship when CCI data are a major component of a paper.

Information on the datasets used in this study was compiled with the help of the Korean Climate Data Navigator (KCDN) (Sim, 2026; https://doi.org/10.5281/zenodo.23300923).

Links¶

• ESA CCI Sea Surface Salinity project page
• CEDA — SSS CCI v5.5 collection
• CEDA — all SSS CCI datasets (by version)
• SSS CCI Product User Guide (PUG v4.1, PDF)
• SSS CCI Product Validation and Intercomparison Report (PVIR v5.1, validates v5.5, PDF)
• ESA CCI Salinity terms and conditions (PDF)
• Poster on the planned C3S sea surface salinity record (2026-05)
• SMAP sea surface salinity on this site (a CCI input mission)
• SMOS sea surface salinity on this site (a CCI input mission)