Science Meetings

Sea Water Permittivity Model and Differences in Sea Surface Salinity Retrieved from SMOS and Aquarius
Dinnat, E. and Le Vine, D.M. (16-Apr-13)

The ESA Soil Moisture and Ocean Salinity (SMOS) mission and the NASA Aquarius instrument share the common scientific objective of mapping the global Sea Surface Salinity (SSS). To that end, they both use radiometers at L-band to measure microwave emission from the sea surface. The principle for the measure of SSS by means of L-band radiometry relies on the change of sea water permittivity with the salt content. Yet, our knowledge on the influence of salt and temperature on sea water permittivity is still imperfect, and it is a potential source of error in the retrieval of SSS. The permittivity models used for L-band radiometry [e.g. Klein and Swift, 1977; Meissner and Wentz, 2004] exhibit differences of less than a percent. Yet, this leads to differences of the order of a few tenths of a Kelvin in brightness temperature (Tb), depending on the temperature and salinity of sea water, and the sensor's incidence angle. The difference in retrieved salinity between these models should be a few tenths of practical salinity unit (psu) most of the time, but it will vary regionally and seasonally according to changes in environmental conditions.

Differences between SSS retrieved by SMOS and Aquarius are mostly between -1 psu and +1 psu, with a significant regional and latitudinal dependence. Part of this difference is expected to be due to the different sea water permittivity models used for the two missions. Aquarius uses a slightly modified version of the model by [Meissner and Wentz, 2004] while SMOS uses the [Klein and Swift, 1977] model. Both models differ in their dependence on salinity and temperature, leading to differences in Tb between +/- 0.3K that will exhibit regional and seasonal patterns. The permittivity model is used at two stages during the salinity retrieval: 1/ to calibrate the instruments by comparing radiometric measurements to simulation using a forward model, and 2/ at the final stage of the retrieval, to compute SSS from retrieved surface Tb. For Aquarius, the calibration involves correcting the measured antenna temperatures (Ta) for a constant bias of the order of a few Kelvin, and for a temporal drift that is composed of an exponential decrease of the order of 1K over 6 months and of 0.2K oscillations with monthly time-scales. In order to assess how much of a contribution the difference in permittivity model has on the SSS difference between SMOS and Aquarius, we reprocess the Aquarius data using the [Klein and Swift, 1977] permittivity model, adjusting the calibration of the Ta's and performing the inversion with the new model.

We will present global maps of the difference between the original Aquarius data and the reprocessed Aquarius data using the same permittivity model as the one used for SMOS. The difference varies at global scale mostly within 0.5 psu, with a few localized larger variations. Significant seasonal variation occurs, particularly at mid and high latitudes. These SSS differences due to permittivity model will be compared to the SSS differences observed between SMOS and Aquarius in order to determine the impact of the permittivity model on the SSS inconsistency between the two instruments. We will focus on a few areas of interest to compare the performances of the permittivity models in reproducing in situ measurements.

Klein, L. A., and Swift, C. T., 1977, An improved model for the dielectric constant of sea water at microwave frequencies. IEEE Transactions on Antennas and Propagation, AP-25, 104-111.

Meissner, T. and Wentz, F. J., 2004, The Complex Dielectric Constant of Pure and Sea Water from Microwave Satellite Observations. IEEE TGARS 42(9): 1836-1849.