Satellite-based T/S Diagrams: a Prospective Diagnostic Tool to Trace Ocean Water Masses
Sabia, Roberto1; Klockmann, Marlene1; Donlon, Craig2; Fernández-Prieto, Diego1; Talone, Marco3; Ballabrera, Joaquim4
1ESA, ITALY; 2ESA, NETHERLANDS; 3SERCO, ITALY; 4SMOS-BEC, SPAIN

Temperature-versus-Salinity (T/S) diagrams are characteristic property-versus-property plots traditionally used to emphasize the mutual relationships between these two variables with the aim of identifying and tracing water masses. Intensively used in oceanography, they basically capitalize on the broad collection of spatial and temporal in-situ measurements, principally from vertical profiles. In contrast, the purpose of this study is to rely on satellite measurements, which, despite the accuracy of the representation, benefit from synopticity and frequent temporal coverage. This study has become feasible thanks to recent satellite missions measuring Sea Surface Salinity (SSS), allowing derivation of spaceborne T/S diagrams through the concurrent availability of Sea Surface Temperature (SST) estimations. This study inherently relates to the horizontal surface section of T/S diagrams, which poses some challenges in interpreting results.

The Soil Moisture and Ocean Salinity (SMOS, [1]) and the Aquarius/SAC-D [2] satellite missions, through their estimation of SSS, provide a means to study, for the first time, the temporal variation of surface water masses on a global scale. The ESA SMOS mission, launched in November 2009, has been in its operational phase since May 2010. An iterative inversion scheme [3] provides SSS estimations (Level-2 product) from passive multi-angular MIRAS brightness temperatures (TB, Level-1 product) by minimizing the difference between measured and modeled TBs, employing constraints on the SST and wind speed auxiliary fields. The mission-prescribed Level-3 SSS accuracy, after adequate spatio-temporal averaging, is 0.1 pss over 2° x 2° boxes in 30 days.
As a parallel effort to SMOS, Aquarius/SAC-D is a US/Argentine combined active-passive mission, launched in June 2011 and operational since December 2011. Aquarius aims to estimate surface salinity fields with an accuracy of 0.2 pss, after temporally averaging over a month [4]. A crucial capability of Aquarius lies is the collocated roughness information provided by its L-band scatterometer, a necessary auxiliary parameter for proper SSS retrieval.

Characterizing the variability of satellite SSS in relation to the existing climatology is key to understanding 1) the unique information provided by SMOS and Aquarius SSS data and 2) the processes governing the distribution and variability of SSS. In [5], T/S diagrams derived from World Ocean Atlas (WOA) climatology are used to obtain surface baseline patterns for comparison to satellite-based T/S diagrams and analysis of geographical differences on a regular 1°x1° resolution grid for the month of September 2011. The approach used is to cluster water masses by appropriately splitting the oceans into seven regions using the 19 world ocean upper-water masses classification of [6]. This methodology allows the segmentation of the T/S maps into several geometric loci, for which differences in each class have been analysed.
Subsequently, in [7], ad-hoc mismatch indices were derived to emphasize the SSS/SST deviations from climatology in terms of the combined modulus and angle (relative under/over-estimation in terms of SSS and SST). Additionally, several months of data were analysed to provide a first dynamic time-series view of the T/S diagrams in each region.

Currently, the T/S diagrams are produced from a growing ensemble of satellite datasets: 1) SSS data distinguished into binned and Optimally-Interpolated (OI) products, namely SMOS L3 and Aquarius L3; and 2) SST data from the Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) and WindSat instruments, as well as the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA, [8]). In addition to WOA climatology, the satellite-based T/S diagrams are compared to ARGO-interpolated In Situ Analysis System (ISAS) fields [9]. Ongoing efforts address the interpretation of the geographical deviations in each region with respect to the baseline, refining mismatch indices and analysing different thresholds. In order to relate these mismatches to identifiable oceanographic structures and processes, additional satellite datasets of ocean currents, evaporation/precipitation fluxes, and wind speed are being super-imposed.
Future work will explore additional SSS and SST datasets, refine the water mass classification, and try to relate the surface T/S signal to the vertical structure and buoyancy-driven ocean circulation processes. Longer time series will help to identify seasonal to interannual variability and shifts or trends in surface water mass formation.
Eventually, comparison with ARGO-based T/S diagrams might provide evidence of SSS biases and errors currently experienced by the satellites (e.g., due to roughness models applied in the SSS retrieval or external noise sources, such as galactic noise, ionosphere, sun glint, or radio-frequency interference (RFI) [10]), and, in a wider context, may provide an alternative means for gaining insights into the oceanic branch of the hydrological cycle.

References

[1] Font, J., A. Camps, A. Borges, M. Martin-Neira, J. Boutin, N. Reul, Y. H. Kerr, A. Hahne, and S. Mecklenburg, "SMOS: The challenging sea surface salinity measurement from space," Proceedings of the IEEE, vol. 98, pp. 649-665, 2010.
[2] Lagerloef, G., "Satellite mission monitors ocean surface salinity, " Eos Trans. AGU, 93(25), 233, 2012.
[3] Zine, S., J. Boutin, J. Font, N. Reul, P. Waldteufel, C. Gabarró, J. Tenerelli, F. Petitcolin, J.L. Vergely, M. Talone, and S. Delwart, "Overview of the SMOS Sea Surface Salinity Prototype Processor," IEEE Trans. Geosc. Remote Sens, 46 (3), 621-645, 2008.
[4] Le Vine, D.M.; Lagerloef, G.S.E.; Torrusio, S.E.; "Aquarius and Remote Sensing of Sea Surface Salinity from Space," Proceedings of the IEEE , vol.98, no.5, pp.688-703, May 2010, doi: 10.1109/JPROC.2010.2040550.
[5] Sabia, R. et al., "Derivation of an Experimental Satellite-based T-S Diagram", Proceedings of IGARSS '12 (International Geoscience and Remote Sensing Symposium), Munich, Germany, July 2012.
[6] Emery, W. J., Water Types and Water Masses, Ocean Circulation, Elsevier science, pp 1556-1567, 2003.
[7] Sabia, R., J. Font, C. Donlon, M. Talone, D. Fernandez-Prieto, E. Bayler, J. Ballabrera, G. Lagerloef, Y. Chao, "Preliminary Results on the Derivation of Satellite-Based T/S Diagrams,AGU Fall Meeting, San Francisco, USA, December 2012.
[8] Donlon, C. J., M. Martin, J. D. Stark, J. Roberts-Jones, E. Fiedler and W. Wimmer, 2011. The Operational Sea Surface Temperature and Sea Ice analysis (OSTIA). Remote Sensing of the Environment. doi: 10.1016/j.rse.2010.10.017 2011.
[9] Gaillard, F., E. Autret, V.Thierry, P. Galaup, C. Coatanoan, and T. Loubrieu , 2009 : Quality control of large Argo data sets. JAOT, Vol. 26, No. 2. 337-351.
[10] Font, J. et al., "SMOS first data analysis for sea surface salinity determination," Int. Journal of Remote Sensing, Vol. 34, Iss. 9-10, 2013.