Table 1. The numbers of composite members used in the analysis
Fig. 1.
Composite of observed Z300 anomaly for entire El Niño events during (a) early winter (ND) and (b) late winter (JFM). Note that (c) and (d) are same as (a) and (b) except for SST anomaly. The contour indicates geopotential height and SST anomalies with 10 m and 0.3°C intervals, from 10 m and 0.15°C, respectively. Light and dark shadings indicate the 90% and 95% confidence levels, respectively.
Fig. 2.
(a) Joint composite maps of the observational Z300 and SST anomalies over the North Atlantic for El Niño years accompanied by the cold TNAL SST anomaly (TNAL index <0) during early winter. The contour indicates the geopotential height and SST anomalies with 10 m and 0.2°C intervals, respectively. Zero contours were omitted. Red shading denotes positive anomalies, while blue shading denotes negative anomalies, which are significant at a 95% confidence level. Figure (b) is the same as (a), but it represents El Niño years with warm TNAL SST anomalies (TNAL index >0). Figures (c) and (d) are the same as (a) and (b), respectively, for late winter data.
Fig. 3.
Same as Fig. 1, but for GFDL CM2.1 coupled model output.
Fig. 4.
Temporal variation of the TNAL index during El Niño years from observation data (black solid line) and model output (grey dotted line). In order to appropriately compare the observation and model results, El Niño events with Niño-3.4> σ were considered only (11 El Niño episodes in the observation data).
Fig. 5.
Same as Fig. 2, but for GFDL CM2.1 coupled model output.
Fig. 6.
Joint composite maps for Z300 anomalies using (a) observation and (b) GFDL 2.1 coupled model output during early and late winters that were accompanied by a cold TNAL SST (TNAL index<−σ) and by no ENSO (−0.5 σ< NDJ Niño-3.4 index <0.5 σ). Contours represent anomalies with 10 m interval. Light and dark shadings indicate the 90% and 95% confidence levels, respectively. The maps in (c) and (d) are the same as (a) and (b), respectively, but for a warm TNAL condition (TNAL index >σ).
Fig. 7.
(a) Scatter plot of the TNAL index against the corresponding NAO index for early winter of each El Niño year in the observation. The TNAL and NAO indices were both normalised. (b) Same as (a) but for late winter. Closed circles indicate stronger El Niño years when the Nino-3.4 exceeds σ.
Fig. 8.
(a) Joint composite maps of the observational surface temperature anomalies for the El Niño years accompanied by the cold TNAL SST anomaly during early (top) and late (bottom) winters. Contours represent anomalies with 1°C intervals from 0.5°C. Light and dark shadings indicate the 90% and 95% confidence levels, respectively. (b) Same as (a) but for El Niño years with a warm TNAL SST anomaly.
Fig. 9.
(a) Joint composite maps of the observational vertical temperature anomaly profile (top), Z300 anomaly (middle) and SST anomaly (bottom) for the El Niño years accompanied by stratospheric warming. For the vertical temperature profile, scaling for the shading is given in the right of figure (°C). The black solid line indicates a significant area at a 90% confidence level. For the Z300 and SST anomalies, contour intervals are 10 m and 0.15°C, respectively, and zero contours were omitted. Light and dark shadings indicate the 90% and 95% confidence levels, respectively. (b) Same as (a) but for the El Niño years without the stratospheric warming. (c) Z300 (upper) and SST (bottom) anomalies for the El Niño years accompanied by the TNAL SST warming (TNAL index >0.5 σ) among the years without the stratospheric warming. (d) Same as (c) but for El Niño years not accompanied by the TNAL SST warming (TNAL index <0.5 σ).
