Abstract
All major ocean tide constituents are aliased into signals with periods less than 90 days from TOPEX/POSEIDON altimetry, except the K-1 constituent. The aliased K-1 has a period of 173 days. Consequently, it might be confounded with height variations caused by the semiannual cycle having a period of 183 days. The correlation between K-1 and the semiannual signal has been investigated both locally and globally using combinations of T/P, ERS-1 and GEOSAT observations. Subsequently, two empirical methods have been investigated to improve the mapping of K-1 from multiple satellites. At high latitudes, where the presence of crossing tracks cannot separate K-1 from the semiannual signal from TOPEX/POSEIDON, the importance of including ERS-1 and GEOSAT observations was demonstrated. A comparison with 29 pelagic and coastal tide gauges in the Southern Ocean south of 50 degrees S gave 5.59 (M-2), 2.27 (S-2) and 5.04 (K-1) cm RMS agreement for FES95.1 ocean tide model. The same comparison for the best empirical estimated constituents based on TOPEX/POSEIDON + ERS-1 + GEOSAT gave 4.32, 2.21, and 4.29 cm for M-2, S-2 and K-1, respectively.
Original language | English |
---|---|
Journal | Progress in Oceanography |
Volume | 40 |
Issue number | 1-4 |
Pages (from-to) | 197-216 |
ISSN | 0079-6611 |
DOIs | |
Publication status | Published - 1997 |