Which Salinity Data are Best for You?

« SMAP Land Correction - Part 2 | Research Insights – Gulf of Mexico »

SMAP Land Correction - Part 3

JPL

Look-Up Table Approach (Level-1B)

  • Generate LUT of the land fraction (fland) as a function of latitude, longitude, and antenna azimuth angle based on SMAP data
  • Generate LUT of land TB climatology for each month (H and V polarizations)
    SMAP TbH clim. ascending
    Example Look-Up Table (LUT) of land TB climatology for H-pol data during the months of January collected by SMAP over 3 years (2016-2018).
  • For each *ocean point within 1000 km of land, compute the average TBl,near value for all land points within 500 km of that *ocean point (see image)
  • All of this is used to calculate the land-corrected TB or TBlc:
    Elements used to calculate TBl
    Elements used to calculate TBl,near of *ocean point at the circle's center.
    Equation
    Where: TBlc = land-corrected TB;
    TBobs = observation of TB at *ocean point; and
    TBl,near = expected contribution of land TB to *ocean point based on place & time
JPL SMAP sea surface salinity map with and without land correction
Below are land correction validation graphs based on HYCOM. Improvement in sea surface salinity (SSS) retrieval accuracy as a function of land fraction and distance from coast. LEFT: Land fraction corresponds to the magnitude of the land signal into the antenna. We see a very significant improvement in the bias and RMS of the retrieved salinity as compared to HYCOM. RIGHT: The odd-looking reversal in trend versus distance to coast can be due to very small islands which can be close but not have significant land contamination.
Land correction validation graphs
« Previous Next »