High-Resolution 3D thermohaline fields derived from in situ and satellite observations (2012)

S.Guinehut, A.-L.Dhomps, G.Larnicol (1), P.-Y.Le Taon(2)

  

* Corresponding author : Stéphanie Guinehut

(1) : CLS Space Oceanography Division, Toulouse France

(2) : IFREMER, Brest France

Full paper: Guinehut S., Dhomps A.-L., Larnicol G. and Le Traon, P.-Y., 2012: High Resolution 3-D temperature and salinity fields derived from in situ and satellite observations, Ocean Science Discuss. 9: 1313-1347, doi:10.5194/osd-9-1313-2012.

To understand, monitor and predict the ocean state, Global Ocean Observation-based Products (GOOP) that combines satellite and in-situ measurements have been developed. Global instantaneous 3-D thermohaline fields are provided at high temporal and spatial resolution, and near real time products are available for the full 1993-2009 time period. The main contribution of this study is to demonstrate that the main components of the global ocean observing system can be integrated efficiently using statistical methods. The first ojective is to provide a global description of statistical relationships that exist between surface and subsurface fields using in situ observations. The second one is to quantify the capacity of such relationships to be used to reconstruct the interannual variability of the 3-D Ocean thermohaline fields together with additional in situ observations.

Data and Method

A 3-D ocean state is obtained by merging only observations from different sources. In-situ T/S fields from the CORA 3.1 dataset, provided by the Coriolis GDAC and including Argo floats, CTDs, XBTs and moorings, are used for the 1993-2009 time period (Cabanes et al., 2011). In-situ historical T/S profiles, build by additioning Argo floats observations available at Coriolis GDAC as of February 2009 to the EN3 dataset (Ingleby and Huddleston, 2007), are used to compute the statistics that related the surface to the subsurface fields. Gridded maps of altimeter Sea Level Anomalies (SLA) are delayed-mode products computed by the SSALTO/DUACS center thanks to an optimal combination of all available satellite altimeters (AVISO,2012). Gridded maps of wednesday fields of Sea Surface Temperature (SST) from the Daily Reynolds L4 analyses of the National Climatic Data Center at NOAA (Reynolds et al., 2007), are used with a 1/4° horizontal resolution for the 1993-2009 time period. ARIVO climatology is used to compute anomalies of the T and S profiles from a climatological monthly means (Gaillard and Charraudeau, 2008). Gridded monthly fields of (T,S) computed from Argo floats and generated by the Scripps Institution of Oceanography are used for validation only.

The first step of the merging method consists in deriving synthetic T and S fields from altimeter and SST observations using a multiple/simple linear regression method and covariances calculated from historical in-situ observations. A comparison with independent dataset is done to validate this first step for the year 2009, and to check the impact of the combination altimeter and SST observations to derive synthetic T and S fields. At the global scale, results show that the temperature bias that existed at all depths when using the ARIVO climatology fields is reduced when using the synthetic estimates (fig.1a) . Mean errors are almost zero for all depths when using the synthetic estimates. Rms error range from 0.5°C at the surface, with a maximum of 0.9°C in the mixed-layer depth, decreasing to 0.15°C down to 1000m. Compared to the use of the climatological estimates, results indicate that 50% to 30% at depth of the temperature fields can be reconstructed from altimeter and SST satellite observations and a statistical method. For salinity, only about 30 to 20% of the salinity can be reconstructed from altimeter observations (fig. 1b).

Figure 1a.  Mean (dotted line) and rms error in reconstructing subsurface T using the ARIVO monthly climatology (red), the synthetic fields (blue) and the combined fields (green). Rms error as percentage of signal variance (calculated from the ARIVO climatology) are also showed on right figure. 60400 independent T profiles of the year 2009 have been used for comparison. 

Figure 1b. Same as figure 1a for subsurface salinity. 55000 independant S profiles of the year 2009 have been used for the comparison.

The second step is a combination of the synthetic estimates with all available in-situ T/S profiles using an optimal interpolation method (Bretherton et al, 1976). The main objective is to correct the large-scale part of the synthetic fields (errors and bias are introduced by the regression method) using the surrounding in-situ profiles.

Results

The interannual variability reconstructed by the observation-based fields is then studied. Yearly zonal averages of the synthetic and combined fields as anomalies from the 2004 to 2009 periods are compared to the SCRIPPS independent estimate (Roemmich and Gilson, 2009).

Figure 2. Yearly zonal averages of the synthetic, combined and SCRIPPS temperatures as anomalies from the 2004 to 2009 periods (in °C).The latitudinal extent of each fields are slightly different. 

For the temperature field, results are very similar in terms of amplitude and geographical position for the synthetic, combined and SCRIPPS estimates (figure 2). At the equator, the signals vary from -0.4°C to +0.4°C with a very clear and strong year to year baroclinic variability. At mid and high latitudes, the amplitudes of the signal are smaller (from -0.2°C to +0.2°C) but their structure is vertically coherent down to 1000m and even deeper between 40 to 50°N in the Atlantic Ocean. The year to year variability in the Southern Ocean shows globally very low amplitude signals (<0.05°C) for the 2004 to 2009 periods, which is very consistent with von Schukmann et al.,2009 results. We demonstrate thus that satellite observations such as SLA and SST, combined with a statistical description of the vertical structure of the ocean are able to reconstruct the interannual variability patterns of the 3-D temperature field, without introducing bias or spurious signals.

The temperature variability of the 17-years from 1993 to 2009 is finally qualitatively described using the combined fields. Globally, as for the 2004 to 2009 periods, the amplitude of the signal varies between -0.4°C to +0.4°C with maximum value up to 1.2°C. Again, a strong interannual variability is observed in the equatorial region with a succession of deepening and outcrop of the main thermocline. A clear long-term warming is observed in the Southern Ocean of up to 0.8°C for the 1993 to 2009 period with a signature down to 1300m, for all latitudes between 50°S and 20°S.

Conclusion

All available observations of the ocean (satellite observations as SLA, SST and in-situ observations as T/S Argo profiles and CTDs) are merged to produce weekly 3D maps of temperature and salinity from the surface to 1500m depth. The observation-based approach developed here provides an improved and more complete estimate of the state of the ocean compared to estimates based on in situ or satellite observations only. It is also very consistent and very complementary to what ocean reanalysis that combines model and observations through an assimilation method give (Stammer et al., 2010). For the future, the relevance and accuracy of this combination estimates will depend strongly on the existence of a complete (satellite+in-situ), homogeneous, and sustainable observations system.

References
  • AVISO : SSALTO/DUACS User Handbook: (M)SLA and (M)ADT near-real time and delayed time products, SALP-MU-P-EA- 21065-CLS, ed. 2.9, 73 pp., 2012.
  • Bretherton, F. P., Davis, R. E., and Fandr y, C. B.: A technique for objective analysis and design of oceanographic experiments applied to MODE-73, Deep-Sea Res., 23, 559–582, 1976.
  • Cabanes, C., Grouazel, A., Coatanoan, C., and Turpin, V.: Coriolis Ocean database for ReAnalysis – CORA3 Documentation, ed. 1.0, 50 pp., (http://www.coriolis.eu.org/Science/Data-and-Products/CORA- Documentation), 2011.
  • Gaillard, F. and Charraudeau, R.: New climatology and statistics over the global Ocean, MERSEA-WP05-CNRS-STR- 001-1A, 2008.
  • Ingleby, B. and Huddleston, M.: Quality control of ocean temperature and salinity profiles - historical and real-time data, J. Mar. Sys., 65, 158–175, doi:10.1016/j.jmarsys.2005.11.019, 2007.
  • Reynolds, R. W., Smith, T. M., Liu, C., Chelton, D. B., Casey, K. S., and Schlax, M. G.: Daily high-resolution blended analyses for sea surface temperature, J. Climate, 20, 5473–5496, 2007.
  • Roemmich, D. and Gilson, J.: The 2004–2008 mean and annual cycle of temperature, salinity and steric height in the global ocean from the Argo program, Prog. Oceanogr. 82, 81–100, 2009.
  • Stammer, D., Köhl, A., Awaji, T., Balmaseda, M., Behringer, D., Carton, J., Ferry, N., Fischer, A., Fukumori, I., Giese, B., Haines, K., Harrison, E., Heimbach, P., Kamachi, M., Keppenne, C., Lee, T., Masina, S., Menemenlis, D., Ponte, R., Remy, E., Rienecker, M., Rosati, A., Schrëoter, J., Smith, D., Weaver, A., Wunsch, C., and Xue, Y.: Ocean Information Provided through Ensemble Ocean Syntheses, in: Proceedings of OceanObs’09: Sustained Ocean Obser vations and Information for Society, edited by : Hall, J., Harrison, D. E., and Stammer, D., Vol. 2, Venice, Italy, 21–25 September 2009, ESA Publication WPP-306, 2010.
  • Von Schuckmann, K., Gaillard, F., and Le Traon, P.-Y.: Global hydrographic variability patterns during 2003–2008, J. Geophys. Res., 114, C09007, doi:10.1029/2008JC005237, 2009.