Skip to content. | Skip to navigation

Laboratoire d’Etudes en Géophysique et Océanographie Spatiales

Personal tools

This is SunRain Plone Theme


You are here: Home / Projects / Geographical Projects / Zones Polaires & Glaciers / Survostral / Scientific Results / 4) Sea surface salinity variations and the freshwater budget in the Southern Ocean.

4) Sea surface salinity variations and the freshwater budget in the Southern Ocean.

by Webmaster Legos last modified Aug 27, 2020 10:48 AM


Data from the SURVOSTRAL thermosalinograph enabled us to specify the seasonal structure of long-term surface temperature and salinity and to validate the climatologies in this region (Chaigneau and Morrow, 2002; Morrow and Kestenare, 2014). Early studies showed that the seasonal temperature cycle is well represented by the climatological data from Levitus, but the climatological salinity cycle is too smooth in the frontal regions, and too variable south of 50°S. These salinity biases are now corrected with the new salinity climatologies based on Argo data (eg ISAS/CORA from Coriolis), although the fronts remain too smoothed in these gridded products, and the coastal structure in the north and south of the sections as not well sampled by Argo data. SURVOSTRAL SSS products retain the highest spatial resolution across the ACC (Figure 4.1). 


Figure 4.1: Left : Position of the mean Survostral line (in white) over laid on the mean SSS from the ISAS gridded product, with the main polar fronts marked.  Right.  Monthly averages of Survostral SSS for the period 1993-2012 from Tasmania (43°S) to Dumont D’Urville, Antarctica (67°S), from Oct to Mar, superimposed. From Morrow & Kestenare, 2014. 

In the Subantarctic region north of the Subantarctic Front, interannual variations in the summer SSS and SST signatures (Figure 4.2) are linked to the latitudinal movements of the Subtropical Front (Morrow and Kestenare, 2014). When this front shifts southward, more high salinity subtropical waters are brought into the domain. Rather than responding to local wind stress forcing, the interannual SSS variability is strongly linked to southward flow from eastern Tasmania (the Tasman outflow), dominated by eddy movements (Pilo et al., 2015). A freshwater budget in the surface layer shows that variations in the local surface freshwater (P-E) forcing make a minor contribution to the SURVOSTRAL SSS signature. There appears to be a regime shift in the surface forcing and the SSS response, before and after the large perturbation in 2001–2002 following the 1999-2000 La Nina event. 

Figure 4.2. Temporal evolution of the monthly mean SSS from December to March of each year (in color), averaged over the latitude band 46.5-51°S in the SAZ. The seasonal mean SSS value, calculated from the DJFM monthly mean SSS is shown in black. 

In the Antarctic Zone south of the Polar Front, surface salinity observations from the SURVOSTRAL programme show that salinity at high latitude decreased by 0.1 during the period 1992-2004; this small decrease was initially linked to an increase in precipitation between 50-60°S during this period (Morrow et al., 2008). Recent analyses over a longer 20-year period covering 1992-2011, and with several products of evaporation and precipitation fluxes have revised this result. The decline in salinity was quite marked in the early 1990s, but since 2002, salinity in the Antarctic Zone has stabilized (Figure 4.3). The 20-year SSS trend is not significant (Morrow and Kestenare, 2014). The different freshwater flux products vary between them, and no product follows the observed salinity evolution.

Lateral advection is a key factor in these interannual variations in the Antarctic Zone (Morrow and Kestenare, 2014). Each year, the path followed by the waters detected along the SURVOSTRAL line varies, either with a direct path from the sea ice zone, or with a longer path further north where surface melt waters have more time to mix with saltier subsurface waters. Thus, the SSS response in the Antarctic Zone is more complex than originally estimated.


Figure 4.3 : Mean SSS evolution between 1993-2012 in the Antarctic Zone 54-57°S, (DJFM), (black line) versus % sea ice concentration in an upstream box [135-142°E; 60-70°S] (grey line drawn) (Morrow & Kestenare, 2014). 

In the Sea Ice Zone (SIZ) south of 60°S, SSS shows distinct seasonal variations, based on 22-years of Survostral data (Morrow and Kestenare, 2017). In the northern sea-ice zone during the warming, melting cycle from October to March, waters warm by an average of 3.5 °C and become fresher by 0.1 to 0.25. In the southern sea-ice zone, the surface temperatures vary from − 1 to 1 °C over summer. The largest SSS range occurs in December, with a minimum SSS of 33.65 near the Southern Boundary of the ACC, reaching 34.4 in the shelf waters close to the coast. The main fronts, normally defined at subsurface, are shown to have more distinct seasonal characteristics in SSS than in SST.

Interannual variations in SSS in the SIZ were more closely linked to variations in upstream sea-ice cover than surface forcing (Figure 4.4). SSS and sea-ice variations show distinct phases, with large biannual variations in the early 1990s, weaker variations in the 2000s and larger variations again from 2009 onwards. The calving of the Mertz Glacier Tongue in February 2010 led to increased sea-ice cover and widespread freshening of the surface layers from 2011 onwards. Summer freshening in the northern sea-ice zone is ~0.05–0.07 per decade, increasing to 0.08 per decade in the southern sea-ice zone, largely influenced by the Mertz Glacier calving event at the end of our time series. Knowledge of these interannual variations and trends is important because a large volume of fresh water in summer can create a barrier layer at the surface, thus limiting vertical exchanges between the cool, fresh surface waters and the warmer deeper layers, contributing to a net warming of the deeper water masses.


Figure 4.4 Top panel : Mean SSS (DJFM) (black line *) in the westward coastal flow in the southern Sea-Ice Zone (64.7°S to coast). In grey, the November averaged sea-ice concentration in the Mertz Glacier box 140-145°E. Red stars indicate Aoki et al (2013) shelf WW CTD SSS anomaly values (their Fig 6d). (Morrow & Kestenare, 2017).
Bottom panel) November averaged latitudinal position of the zonal zero wind stress in the box 140-145°E (in violet), versus the December mean sea-ice coverage in the 135-140°E box (in grey). (Morrow & Kestenare, 2017).



Document Actions