| Abstract |
Analysis of Geosat sea level data over the period November 1986-November 1987 has evidenced long, low-frequency equatorial wave propagations over most of the equatorial Pacific basin (Delcroix et al.,1991). The succession of downwelling Kelvin, upwelling Kelvin, and upwelling Rossby waves has raised a series of questions. Is the observed upwelling Rossby wave due to a reflection process on the eastern boundary, or due to wind forcing near the equator? Why is there no evidence of a reflection of the downwelling Kelvin wave? Is this wave sequence a feature of the 1986-1987 El Nino phenomenon or is it to be found in the mean seasonal cycle? We used a linear model forced by the FSU wind stress to study these questions. The results, during the period November 1986-November 1987, show the successive downwelling/ upwelling Kelvin waves and upwelling Rossby wave as in the Geosat observations. Another test was made, using only the part of the forcing responsible for creating Kelvin waves, in order to evaluate the portion of the signals that was due to the reflection process and the part due to local forcing. It appears that wind forcing favourable to upwelling in the eastern part of the equatorial basin is the main mechanism which both generates the upwelling Rossby wave and inhibits the reflection of the downwelling Kelvin wave. However, the reflection of the upwelling Kelvin wave contributes to the upwelling Rossby wave signal, and makes it visible all the way from the eastern to the western sides of the equatorial Pacific. Results of a climatological run show that the successive downwelling/upwelling Kelvin and Rossby wave propagations are present in the mean seasonal cycle. However, the corresponding model signals, as well as the wind forcing anomalies, show a larger amplitude in 1986-1987. |
