Evidence That Sea State Bias is Different for Ascending and Descending Tracks

Abstract

Jason-1 and TOPEX/Poseidon (T/P) measured sea-surface heights (SSHs) are compared for five regions during the verification tandem phase. The five regions are of similar latitude and spatial extent and include the Gulf of Mexico, Arabian Sea, Bay of Bengal, and locations in the Pacific and Atlantic Oceans away from land. In all five regions, a bias, defined as Jason SSH—TOPEX-B SSH, exists that is different for ascending and descending tracks. For example, in the Gulf of Mexico the bias for ascending tracks was −0.13 cm and the bias for descending tracks was 2.19 cm. In the Arabian Sea the bias for ascending tracks was −2.45 cm and the bias for descending tracks was −1.31 cm. The bias was found to depend on track orientation and significant wave height (SWH), indicating an error in the sea state bias (SSB) model for one or both altimeters. The bias in all five regions can be significantly reduced by calculating separate corrections for ascending and descending tracks in each region as a function of SWH. The correction is calculated by fitting a second-order polynomial to the bias as a function of SWH separately for ascending and descending tracks. An additional constraint is required to properly apply the correction, and we chose to minimize the sum of the TOPEX-B and Jason-1 root-mean-square (rms) crossover differences to be consistent with present SSB models. Application of this constraint shows that the correction, though consistent within each region, is different for each region and that each satellite contributes to the bias. One potential source that may account for a portion of the difference in bias is the leakage in the wave forms in TOPEX-B due to differing altitude rates for ascending and descending tracks. Global SSB models could be improved by separating the tracks into ascenders and descenders and calculating a separate SSB model for each track.

Keywords:

• altimetry
• Jason
• sea state bias
• TOPEX/Poseidon

Acknowledgments

This work was supported by the Office of Naval Research as part of the Naval Research Laboratory's basic research project “Slope to Shelf Energetics and Exchange Dynamics (SEED)” under Program Element 0601153N (NRL-SSC contribution NRL/JA/7330/04/9). The authors would like to thank Robert Leben and Brian Powell for supplying their MATLAB code for calculating the filter coefficients. Thanks are also extended to Phil Callahan for helpful background information related to the wave form leakage problem in TOPEX-B.