Characterization of Kelvin Waves in the Equatorial Pacific

Adam Rydbeck, US Naval Research Laboratory, Washington, DC, United States, Tommy G Jensen, US Naval Research Laboratory, Ocean Sciences Division, Stennis Space Center, MS, United States and Maria K. Flatau, Naval Research Laboratory, Monterey, CA, United States
Abstract:
This study develops a new technique for the objective identification of subseasonal Kelvin waves in the equatorial Pacific Ocean, including their phase and amplitude, that is based on satellite‐derived sea surface height anomalies. This methodology is then utilized to formulate composites of intraseasonal Kelvin waves to reveal important patterns of wave evolution across the Pacific Ocean. Kelvin waves are an important subseasonal modulator of upper‐ocean thermal and kinetic energies that can rectify onto longer timescales via the Bjerknes feedback mechanism. We observed that the waves maintain an average phase speed of 2.55 m/s, with maximum of 2.73 m/s when the downwelling portion of the wave is near 150°W and minimum of 2.35 m/s near 175°W. Kelvin waves attain their maximum amplitude at 150°W, the region where intraseasonal surface wind forcing strongly relaxes, and the waves transition from a forced to a freely propagating mode. The dominant processes controlling the mixed‐layer temperature also shift near 150°W. Intraseasonal SST anomalies associated with the waves have maximum amplitude of +/−0.25 °C at 140°W, with much stronger anomalies, +/−1.4 °C, along the thermocline at 155°W. Waves strengthen the eastward equatorial undercurrent by 76% while shifting its maximum 3,500 km to the east. The El Niño Southern Oscillation exerts considerable control over the vigor of Kelvin waves, with statistically significant increases of upper ocean kinetic energy during El Niño versus La Niña periods. For the benefit of the forecasting community, a counterpart intraseasonal Kelvin wave index is developed that avoids the use of forward time filtering such that wave phase and amplitude are available in real time.