Quality assessment of spaceborne sea surface salinity observations over the northern North Atlantic

• Published in: Journal of Geophysical Research: Oceans
• Main Authors: Köhler, Julia, Sena Martins, Meike, Serra, Nuno, Stammer, Detlef
• Format: Article in Journal/Newspaper
• Language: English
• Published: AGU (American Geophysical Union) 2015
• Subjects: Greenland; North Atlantic
• Online Access: https://oceanrep.geomar.de/id/eprint/29809/; https://oceanrep.geomar.de/id/eprint/29809/1/K-hler_et_al-2015-Journal_of_Geophysical_Research__Oceans.pdf; https://doi.org/10.1002/2014JC010067

Spaceborne sea surface salinity (SSS) measurements provided by the European Space Agency's (ESA) “Soil Moisture and Ocean Salinity” (SMOS) and the National Aeronautical Space Agency's (NASA) “Aquarius/SAC-D” missions, covering the period from May 2012 to April 2013, are compared against in situ salinity measurements obtained in the northern North Atlantic between 20°N and 80°N. In cold water, SMOS SSS fields show a temperature-dependent negative SSS bias of up to −2 g/kg for temperatures <5°C. Removing this bias significantly reduces the differences to independent ship-based thermosalinograph data but potentially corrects simultaneously also other effects not related to temperature, such as land contamination or radio frequency interference (RFI). The resulting time-mean bias, averaged over the study area, amounts to 0.1 g/kg. A respective correction applied previously by the Jet Propulsion Laboratory to the Aquarius data is shown here to have successfully removed an SST-related bias in our study area. For both missions, resulting spatial structures of SSS variability agree very well with those available from an eddy-resolving numerical simulation and from Argo data and, additionally they also show substantial salinity changes on monthly and seasonal time scales. Some fraction of the root-mean-square difference between in situ, and SMOS and Aquarius data (approximately 0.9 g/kg) can be attributed to short time scale ocean processes, notably at the Greenland shelf, and could represent associated sampling errors there.