More recently White and Alan (1999) found a similar eastward propagating wave in covarying SLP and SST anomalies traveling across the Indian, Pacific, and Atlantic oceans on biennial timescales.The called this the global biennial wave. It too was strongly modulated by interdecadal variability.
The physics of these global scale standing and propagating climate change phenomenon are still largely unknown, with new discoveries being made every day. Research is hampered by the lack of basic oceanic and atmospheric datasets with which to examine heat and vorticity budgets in both the tropical global upper ocean and lower troposphere. Presently the global ocean observing system (GOOS) consists of temperature profiles being collected over the tropical Pacific ocean by the TOGA TAO thermistor chain array and over the global tropical ocean by volunteer observing ships (VOS) deploying expendable bathythermographs (XBT's). However, these nascent oceanographic networks are woefully inadequate. A global ocean observing system has been proposed recently (called the ARGO program) to correct the inadequacy of upper ocean sampling by deploying 3000 or so profiling floats over the global ocean. Each float in this proposed ARGO program are expected to profile temperature and salinity in the water column down to 1000 m every 10 days, with a mean float separation of ±2° latitude-longitude grid and a mean lifespan of 5 years. This is an international program and is expected to operate under the auspices of the IOC and the WMO.
2. Zonal Wavenumber-Frequency Spectrum of Monthly SST Anomalies in the Indo-Pacific Ocean from 1900 To 1997
We begin by examining the zonal wavenumber-frequency spectrum for monthly GISST SST anomalies (Folland and Powell, 1994; Parker et al., 1994) along the equator across the Indo-Pacific ocean for 98 years from 1900 to 1997 (top, Figure 1). Because zonal wavenumber-frequency spectra separate eastward and westward propagating wave energy, they also allow for separation of standing wave and propagating wave energy density (middle and bottom, Figure 1) (von Storch and Zwiers, 1999, pages 241-250). Spectral peaks in zonal wavenumber-frequency spectra occur at periods of 2.3, 3.0, 3.6, 5, 7, 10, and 24 years (top, Figure 1), with contours of peak spectral energy density significantly different from adjacent contours at the 90% confidence level (Jenkins and Watts, 1968, pages 77-89). In the propagating wave spectrum (middle, Figure 1) eastward propagating signals dominate westward propagating signals at ENSO periods of 3.0, 3.6, and 5 years, while westward propagating signals dominate eastward propagating signals at decadal and interdecadal periods of 7, 10, and 24 years. In the standing wave spectrum (bottom, Figure 1), significant peaks are also found at ENSO periods of 3.6 and from 5 to 7 years. All ENSO, decadal, and interdecadal signals are associated with zonal wavelengths greater than 120° of longitude, indicating that they are global in extent (that is, they are associated with global zonal wavenumber 1, 2, and 3). These global-scale eastward propagating waves of 3 to 5 year period can be attributed to a global propagating ENSO wave while the global-scale standing waves of 3 to 7 year period can be attributed to the Southern Oscillation (or the global standing ENSO wave). In the propagating wave spectrum an eastward propagating signal is also found at the biennial period of 2.3 years, not observable in the standing wave spectrum.
