An impact of the phase of ENSO on the interaction of midlatitude Rossby waves and equatorial waves

The interannual variability due to the phase of ENSO in the behaviour of equatorial waves is being studied using 19 years of ERA data to ascertain whether equatorial waves are modified by interaction with ENSO and midlatitude disturbances.

Figure 1 shows a wavenumber-frequency power spectrum analysis of 200 Pha meridional wind in the extratropics (Fig.1a) and tropics (Fig.1b) for 1979-98 Northern Hemisphere winter. Superimposed lines in Fig.1a are the theoretical dispersion curves of midlatitude barotropic Rossby waves. The signature of Rossby waves with both eastward and westward phase speed in midlatitudes and their penetration into the tropics is evident. It should be noted that as the data are sampled once per day, the high-frequency westward propagating waves seen in Fig.1a are actually a result of aliasing and these values should really be considered as a continuation of the eastward signatures from 0.5 to 0.8 cpd. The low frequency westward Rossby waves (Fig.1a) may be the signature of blocking events which tend to retrogress. Propagation features of nonstationary Rossby waves were discussed in Yang and Hoskins (1996). The evidence of midlatitude Rossby waves penetrating into equatorial regions is specially true for the eastward moving waves, as equatorial eastward moving Kelvin waves would not appear in meridional winds. For westward moving waves, the signature in Fig.1b could result from midlatitude Rossby wave propagating into the tropics and/or from equatorial westward moving waves. The different equatorial response to higher-latitude forcing with eastward and westward phase speed has been shown in Hoskins and Yang (2000).

Figure 1. Zonal wavenumber-frequency power spectrum of the 200-hPa meridional wind anomaly v' (departure from monthly mean) in the Northern Hemispheric winter season from Nov. to Apr. The power is divided by a background power and the standing wave part is removed. The power is averaged over the 19 winter seasons from 1979 to 1998 and over the latitudes (a) 17- 51°N, and (b) 15°N - 15°S. Superimposed lines are the dispersion curves of the midlatitude barotropical Rossby waves. Solid line: U=30ms^-1, l=3 and =3.0*10^-11 m^-1s^-1. Dotted line: U=35 m^s-1, l=2, and =1.6*10^-11 m^-1s^-1. Units are m² s^-2

A number of studies show that Rossby waves tend to propagate through an equatorial westerly duct in the upper troposphere (e.g., Webster and Holton 1982). In the Northern Hemisphere winter there are two such westerly ducts over the equatorial eastern Pacific and Atlantic regions. To examine if ENSO events have an influence on the westerly wind duct and associated midlatitude Rossby wave activities, Figure 2 gives 200 hPa zonal winds for the recent 4 El Nino and 3 La Nina winters and the difference between them. A global difference, which interestingly is nearly symmetric about the equator, can be seen from Fig.2c. It is notable that the Atlantic westerly duct is much stronger in El Nino winter than in La Nina winter. In contrast, the Pacific westerly duct tends to be much weaker.

Figure 2. 200-hPa zonal wind for (a) four El Nino (1982/83, 1986/87,1991/92 and 1997/98) and (b) three La Nina (1984/85, 1988/89 and 1995/96) winter (Nov. to Apr.) mean. (c) the difference between (a) and (b). Units are ms^-1.

The impact of the phase of ENSO on the penetration of extratropical Rossby waves is demonstrated in Fig 3. Figures 3a to c show the geographical distribution of the standard deviation of eastward moving Rossby waves in El Nino and La Nina winters and the difference between them. As excepted the strongest Rossby wave activity occurs in the extratropics, associated with jet wave guides and storm tracks. Dramatically different behaviour in the activity of Rossby waves penetrating into the equatorial regions is evident, associated with changes in the upper tropospheric zonal flow. In El Nino (La Nina) winters more Rossby waves penetrate into the equatorial Atlantic (eastern Pacific) westerly duct. In addition to the difference in equatorial westerly ducts, there is also a significant difference in the subtropics-midlatitude regions for both the hemispheres at longitudes 180-360o. This variability could be a mixed feature of in situ influence of zonal winds and a remote response to the ENSO-associated heating anomaly in the tropical Pacific.

Figure 3. Geographical distribution of the standard deviation of the eastward and westward moving waves presented by the 200-hPa meridional wind anomaly v' (departure from monthly mean). (a) for eastward moving waves averaged over four El Nino winters, (b) for eastward moving waves averaged over three La Nina winters, (c) the difference between (a) and (b), (d) the same as in (c) but for the westward moving waves. Eastward moving waves are for zonal wavenumber k from 5 to 10, period from 3 to 30 days. Westward moving waves are for k from -2 to -8 and period from 4 to 30 days. Units are ms^-1.

Similar plots to Fig.3a and b but for the westward moving waves show generally smaller standard deviations than those of eastward moving waves, which is consistent with the fact that most synoptic systems in midlatitudes move eastwards with the ambient westerly. However the difference between El Nino and La Nina winters in the eastern Pacific westerly duct (Fig.3d) is as significant as that for eastward moving waves (Fig.3c). Nevertheless, it is not immediately clear that the difference shown in Fig.3d is a result of a weaker influence of midlatitude Rossby waves and/or less westward equatorial waves over the equatorial eastern Pacific in the El Nino winters. Research is continuing.

References