Analysis of vertically propagating convectively coupled equatorial waves using observations and a non‐hydrostatic Boussinesq model on the equatorial beta‐plane

The Spectrum of Convectively Coupled Kelvin Waves and the Madden–Julian Oscillation in Regions of Low-Level Easterly and Westerly Background Flow

“…This result suggests that background westerlies occur there for roughly half of the seasonal cycle, although ENSO and the Indian Ocean dipole also influence the presence of westerlies in that region, leading to variability in their timing relative to the seasonal cycle from year to year. This geographical distribution of counts of westerly background wind data points is remarkably similar to the region of greatest OLR variance in the traditional MJO band of the wavenumberfrequency domain (Wheeler et al 2000;Roundy and Frank 2004a). Although not shown here, these regions of low-level westerly wind migrate with the seasonal cycle and interannual variability of the warm pool, and they tend to collocate seasonally with regions of enhanced variance in OLR anomalies filtered for the MJO (Roundy and Frank 2004a).…”

“…This geographical distribution of counts of westerly background wind data points is remarkably similar to the region of greatest OLR variance in the traditional MJO band of the wavenumberfrequency domain (Wheeler et al 2000;Roundy and Frank 2004a). Although not shown here, these regions of low-level westerly wind migrate with the seasonal cycle and interannual variability of the warm pool, and they tend to collocate seasonally with regions of enhanced variance in OLR anomalies filtered for the MJO (Roundy and Frank 2004a). Thus, assessment of the OLR spectrum with focus on these regions is likely to yield more insight about the spectral characteristics of the MJO than would the spectrum of OLR integrated over the full global tropics.…”

“…Yet, composites of the MJO in the longitude-time lag domain suggest that the propagation characteristics of the MJO depend loosely on geography, and integration of the Fourier transform over different regions of the world would thus contribute different amounts of power to different parts of the spectrum. Results of Hendon and Salby (1994), Roundy and Frank (2004b), and others suggest, for example, that the leading signals in the MJO band would project well onto zonal wavenumber 2 over the warm pool, but onto wavenumber 1 over the Western Hemisphere. That Western Hemisphere projection would occur because most of the hemisphere tends to have anomalous negative or anomalous positive OLR anomalies at the same time in association with eastward radiation of MJO signals across the region.…”

“…This project applies zonal wavenumber-frequency wavelet analysis following Roundy and Janiga (2012) and Roundy (2012) to assess the activity at particular zonal wavenumbers and frequencies at particular geographical regions and at particular times. Kikuchi and Wang (2010) and Wong (2009) …”

“…Kelvin waves are widely recognized as a leading signal among the population of modes that comprise the subscale anatomy of the MJO. Recently, Roundy (2012) applied wavelet analysis in the wavenumber-frequency domain together with linear regression, demonstrating that the structural characteristics of Kelvin waves and the MJO occur as a continuum with dry Kelvin waves at one extreme and the MJO at the other. His results show that signals at progressively lower frequencies from the Kelvin band to the MJO band appear as hybrid patterns, with the low-level geopotential anomalies sliding continuously westward relative to rainfall and low-level westerly wind anomalies as the amplitude of the associated outgoing longwave radiation (OLR) anomalies increases.…”