This paper has two main parts. First, it is shown that if the dynamics is simplified by assuming a one-mode, low-frequency (or long-wave) model, the whole field of anomalous motion in the tropical region can he deduced from the time series of sea-level anomaly at the eastern boundary. This is because the amplitudes of all the equatorial waves are proportional to sea level at this longitude, and the amplitudes at other longitudes can he deduced by integrating along characteristics using the method of Gill and Clarke (1974). The integration proceeds forward in time for the planetary waves and backward in time for the Kelvin wave. The calculation requires a knowledge of wind stress anomalies, although the wind effect only becomes significant after integration to large distances from the eastern boundary. The western boundary is irrelevant to the calculations. The technique is applied to estimate zonal surface current anomalies in the equatorial Pacific for the period 1971-73 spanning a major El Niño. Mean zonal surface current anomalies within ±5° of the equator are deduced to have reached their largest values (of ∼0.5 m s−1) near the date line in early 1972, when they were eastward, and in early 1973, when they were westward.
The second part of the paper uses gm reconstructed anomaly field of the first part to examine anomalies in the heat balance of the surface layer of the equatorial Pacific. The striking result here is that surface temperature anomalies in the central Pacific are well reproduced in both amplitude and timing it caused solely by horizontal advection by the zonal current anomaly. Other mechanisms give much smaller amplitude and the wrong timing. The resultant picture is that warm anomalies in the central Pacific result from huge anomalous eastward movements of warm water from the west Pacific.
The heat balance near the eastern boundary is also discussed, and two possible mechanisms seem capable of reproducing a warm anomaly with the correct strength at the right time. One is anomalous poleward advection of warm water along the coast. The other is upwelling of anomalously warm water near the coast and the spreading of this warm anomaly by advection with the mean flow.