Monthly anomalies at each grid point have been standardized in time. (middle) Standing wave spectrum computed by plotting the minimum spectral energy density estimate at each frequency and zonal wavenumber magnitude in the two quadrants of the zonal wavenumber-frequency spectrum in the top panel. (bottom) Propagating wave spectrum computed by differencing spectral energy density estimates at each frequency and zonal wavenumber magnitude in the two quadrants of the zonal wavenumber-frequency spectrum in the top panel. Spectral energy density contours are 105℃2/(mon-1km-1), significantly different from adjacent contours at the 90% confidence level (Jenkins and Watts, 1968, pages 77-89). Shading is for effect.
Figure 2a. (Top) Zonal wavenumber-frequency spectra (Bendat and Piersol, 1986, pages 361-424) of NCEP monthly SLP and ST anomalies computed from time-longitude diagrams extending along 10°N latitude and circling the globe over the 48 years from January 1950 to December 1997. Spectral energy density contours are 106hPa2/(mon-1km-1) for SLP and 105℃2/(mon-1km-1) for ST, significantly different from adjacent contours at the 90% confidence level (Jenkins and Watts, 1968, pages 77-89). Shading is for effect. (Bottom) Animation sequences of maps for standardized interannual SLP and ST anomalies extending over 5 years from January 1984 to January 1989. Each map in the animation sequence extends zonally around the globe in the latitude band from 40°S to 40°N. These interannual anomalies have been band-pass filtered in time for periods ranging from 3 to 6 years, then standardized in time (see Data and Methods), Individual maps are 6 months apart. The sense of propagation comes from following anomalies of similar sign from one map to the next in each time sequence. Yellow-to-red (blue) colors indicates positive (negative) anomalies, with contours at intervals of 0.4 standard deviations.
Figure 2b. Animation sequences of maps displaying standing and propagating wave components of reconstructed interannual SLP and ST anomalies from the dominant CEOF mode, the latter computed for 24 years from 1973 to 1996, with each map extending around the globe from 40°S-40°N. (top) Animation sequences for standing wave components consist of January 1985 and January 1987, representing extreme amplitudes during La Nina and El Nino respectively. (bottom) Animation sequences for propagating wave components extend over 5 years from January 1984 to January 1989 in 6 month increments. The sense of propagation comes from following anomalies of similar sign from one map to the next in these animation sequence. Yellow-to-red (blue) colors indicates positive (negative) anomalies, with contours at intervals of 0.5 hPa for SLP anomalies and 0.5℃ for ST anomalies.
Figure 3a. Animation sequences of maps for standardized biennial SLP and ST anomalies extending over 2.5 years from January 1981 to July 1983. Each map in the animation sequence extends zonally around the globe in the latitude band from 40｡? to 40｡?. These biennial anomalies have been band-pass filtered in time for periods ranging from 12 to 36 months, then standardized in time (see Data and Methods). Individual maps are 3 months apart. The sense of propagation comes from following anomalies of similar sign from one map to the next in each time sequence. Yellow-to-red (blue) colors indicates positive (negative) anomalies, with contours at intervals of 0.4 standard deviations.
Figure 3b. Animation sequences of maps displaying standing and propagating wave components of reconstructed biennial SLP and ST anomalies from the dominant CEOF mode, the latter computed for 24 years from 1973 to 1996, with each map extending around the globe from 40｡?-40｡?. (top) Animation sequences for standing wave components consist of January 1982 and January 1983, representing extreme amplitudes. (bottom) Animation sequences for propagating wave components extend over 2.5 years from January 1981 to July 1983 in 3 month increments.
