Geophysical model functions for the sea state bias in satellite altimetry

The sea state bias in altimeter measurements of sea surface height have been investigated by many authors, resulting in several proposed algorithms of the form SSB=εH where H is the significant wave height (SWH) and ε is a non-dimensional function of wind and wave parameters. All known algorithms are presently rated. The most popular, linear model of the form ε=a₀+a₁U is shown to yield an improvement over the simplest algorithm with a constant ε. A three-parameter form ε=a₀+a₁U+a₂H produces even better results. However, the accuracy of the polynomial models is below that obtained with a two-parameter, physically-based model relating ε to the pseudo-wave-age ξ: ε=Mξ^-m. The ξ is estimated using altimeter wind U and SWH: ξ≈A(gH/U)^V where A and v are constants. By analyzing the performance of different models we conclude that further progress can hardly be achieved by raising the degree of the polynomial for ε(U,H). Physically-based approaches employing a small number of adjustable parameters and theoretically justified non-dimensional combinations of external wind and wave factors appear to be more promising