Click here to return to SIL home page
Click here to return to SIL home page
The varying proportions of 18-O to 16-O in sea water provide an oceanographic tracer like salinity, but whereas salt is a tracer for the oceanic fluid the isotopic composition is a tracer specifically for the water component of that fluid. The ratio of the oxygen isotopes in sea water is a conservative property altered by the following processes:
1) Evaporation/precipitation: Due to the slightly higher vapour pressure of the molecules containing the lighter oxygen atom, water vapour is isotopically lighter than the water from which it is derived.
2) The mixing of water masses with different isotopic composition.
A further process that affects the ratio of delta 18-O to salinity is the formation of sea-ice. Delta 18-O is a powerful tracer in polar regions as sea-ice formation and melting cause large changes in salinity but leaves delta 18-O almost unchanged, thus decoupling the two tracers.
Samples for delta 18-O analysis were collected in February and March 1993 during the deployment cruise of the Antarctic Deep Outflow eXperiment (ADOX), cruise 200 on board RRS Discovery. We discuss here results from a hydrographic section across the Princess Elizabeth Trough (PET: Dickson, 1993, Fig. 1)
Results are expressed using the delta notation where:
The reference is Vienna Standard Mean Ocean Water (VSMOW).
The temperature-salinity diagram (Fig. 1a) shows four water masses in the PET. The Surface Water (SW) mixes to a Winter Water (WW) layer at about 100 m. Below this lies warmer (2 degrees centigrade) and saltier (34.656 psu) Warm Deep Water (WDW) of Circumpolar origin. The Bottom Water (BW) in the PET has T and S of -0.43degrees centigrade and 34.669 psu.
The potential temperature-delta 18-O diagram (Fig. 1b) shows that the warm surface water at station 12358 has the same delta 18-O value as the much cooler WW layer underneath. There is a linear mixing line from the WW to the warm and isotopically heavier WDW and from there linear mixing to the cooler and lighter BW. The BW temperature and delta 18-O values are intermediate between the WW and WDW.
The salinity-delta 18-O diagram (Fig. 1c) shows a large variation in salinity from the SW to WW (33.1 - 34.4 psu) with little variation in delta 18-O. Thus this SW is warmed WW plus sea ice meltwater. The WW delta 18-O value decreases southward through the Trough from -0.3 ‰ at station 12349 to -0.4 ‰ at station 12358. The salinity increases by approximately 0.3 psu in the WW over this distance. The WDW has the highest salinity and delta 18-O values; there is a tight mixing line between WDW and the BW.
The Bottom Water (BW) has a delta 18-O value of -0.251 +/- 0.004 ‰. This water has T-S characteristics close to that of classical AABW (-0.4 degrees centigrade and 34.66 psu). There are three options for the source of the BW in the PET:
(i) Weddell Sea outflow (further mixing of WSBW with WDW) (ii) westward transport of AABW formed in the Ross Sea, or (iii) local formation occurring seasonally along the Antarctic continental shelf.
Waters with T-S characteristics of AABW found in the Enderby plain and Crozet-Kerguelen trough have delta 18-O values of -0.288 +/- 0.010 ‰. Thus the PET BW is an isotopically heavier water mass quite distinct from the Weddell Sea outflow. The incorporation of ice-shelf water in the formation of WSBW which mixes with WDW to form AABW would cause the lighter isotope value. Similarly, BW from the Ross Sea would be expected to be lighter due to the ice-shelf water component which is of the same range in the Ross Sea as the Weddell Sea.
Here we consider (iii) local BW formation. Assuming that, in winter, the water column temperature throughout the mixed layer would have cooled to -1.9 degrees centigrade, the relative proportions of SW and WDW required to form BW with the observed temperature of -0.43 degrees centigrade from mixing of WDW (2.00 degrees centigrade) entering from the north and SW (-1.90 degrees centigrade) are 38% and 62% respectively. Using these ratios and the measured values of delta 18-O for WW and WDW we predict the BW delta 18-O = -0.249 ‰ which agrees well with the observed value of -0.251 ± 0.004 ‰.
The required salinity of the SW would be 34.678 psu, an increase of less than 0.3 psu on the salinity of the WW found at 150 m at station 12358. This increase would require the formation of approximately 1.2 m of sea-ice, a reasonable amount at this latitude.
The main findings arising from this work are;
1. From delta 18-O results the AABW in the Princess Elizabeth Trough is clearly a water mass distinct from AABW formed from the Weddell basin overflow. The more positive delta 18-O values for PET bottom water indicate that it does not have a significant glacial melt ater component. This is consistent with the view that there are insignificant amounts of Weddell Basin deep and bottom water passing through the PET, and that the deep current is westward through this basin.
2. The classical process of local formation via seasonal deep convection is demonstrated by two independent means requiring only the assumption that the temperature of the surface mixed layer would be close to freezing. The small fractionation of 18-O during sea ice formation allows the data obtained from samples collected in summer to be used to study processes occurring during the winter when the region is inaccessible due to sea-ice cover. The BW found in the PET can be formed from deep convection involving only WDW and winter SW, the relative amounts being about 60% and 40% respectively.