Freshwater budget and salinity variability in the subpolar and subtropical gyres of the North Atlantic

by Monika Rhein and Ilaria Stendardo

In the first phase of this project we used the GEM (Gravest Empirical Mode) technique in order to create high-resolution salinity data in the upper 700 m of the water column from 1993 to 2012. The method uses the first observed mode between T/S profiles and dynamic height in order to parameterize temporal and spatial high-resolution salinity data as a function of dynamic height from satellite altimetry. Therefore the data derived from this technique has the same resolution as the satellite Aviso product, which is daily data projected onto a 1/4° Cartesian grid from the beginning of the altimetry era, 1993, until 2012. To be able to calculate the GEM, the Sea Level Anomaly (SLA) needs to be dominated by baroclinic signal of the water column and a relationship between temperature and salinity is necessary. That means not all the regions in the ocean are suitable for this technique. At the beginning of the first phase of this project, regions in the North Atlantic were selected by looking where the method was working. We were able to define a total of 7 regions mainly distributed along the pathway of the North Atlantic Current (NAC). We validated the method by using several statistical approaches and by comparison with salinity data from Argo and CTD profiles in order to assess the quality of the satellite derived salinity data. After this careful validation, we had to exclude all data below 700 m of the water column. We used the satellite derived salinity data to reconstruct salinity trends from 1993 to 2012 for each region in the upper 700 m of the water column (Figure 1). Thanks to the collaboration with the WP 2.1 “Atlantic surface freshwater fluxes in the HOAPS satellite climatology”, we used the new released of the HOAPS 3.3 climatology to calculate the freshwater fluxes from 1993 to 2012 (Figure 2). The method and validation together with the analysis of the salinity trend over the last 20 years in the North Atlantic are summarized in the paper Stendardo et al., 2016.

Goals for the second phase of this project are:

  • Extend the analysis also in the subtropical regions, when and where this is possible. Also we would like to include some of the regions in the North Atlantic that were excluded at the beginning, for example the Northeastern Atlantic or the Labrador and Irminger Seas. In the last two cases we might need to use only observations since SLA in those regions is not dominated by baroclinic signals.

  • Study the evolution and variability of surface and subsurface structures of fronts, meanders and eddies, calculate salinity content anomalies in eddies, follow salinity anomalies along their pathways and assess the role of the involved processes.

  • Calculate time series of eddy salinity transports in the upper 700 m, especially into the subpolar western Atlantic and across the Mid Atlantic Ridge into the eastern Atlantic.

  • Analyse the monthly to multiannual variability in the regional salinity contents, together with freshwater fluxes and involved processes using observations (Argo, CTD, altimeter), models and results from the GECCO2 synthesis (WP 3.1) and from the updated HOAPS climatology (WP 2.1). Calculate content changes and influence of surface freshwater fluxes for the upper 700 m and separately for the mixed layer.