two periods. The Arctic was completely dominated by negative trends, and the
Antarctic was partially dominated by positive trends in the period of 1978–2002,
while the opposite was generally true in the period of 1958–1977 with the Arctic
partially dominated by positive trends and the Antarctic completely dominated by
negative trends.
For the troposphere, a pronounced positive response to ENSO is observed over
the tropical eastern Pacific (Fig. 4.6a, b), which is opposite to its counterpart in the
stratosphere. This positive ENSO response in the tropical troposphere is present in
the two study periods, while the wavelike response occurs in both middle-high
latitudes. But the structure of the wavelike response shows a significant difference
in the two periods. It implies a different impact of the ENSO events on the
temperature variability over the middle-high latitudes in the two periods.
Compared to the stratosphere, the tropospheric response to QBO was significantly reduced (Fig. 4.6c, d), especially in the tropical areas. The wavelike response
is again identified over the middle-high latitudes although the area exceeding the
statistical significance test at the 95% level is markedly compressed. In addition, the
different responses in the two periods are observed over both middle-high latitudes.
Finally, the tropospheric response to the stratospheric aerosols shows (Fig. 4.6e, f)
an opposing pattern to the one in the tropics (Fig. 4.5e, f). The stronger response
appears over both middle-high latitudes. The amplitude of the response in the period
of 1958–1977 is much higher than its counterpart in the period of 1978–2002. Note
that the significant negative contribution from stratospheric aerosols, which are
impacted via volcanic eruptions, was identified over most areas.
To summarize, the linear trend of the lower stratospheric temperature (Fig. 4.1)
and the strongest temperature anomalies occur repeatedly over the Arctic, but its
sign is positive in 1979–2002 and negative in 1958–1978. The opposite response to
solar cycle forcing (Fig. 4.3) is only observed over the northern middle-high
latitudes. The response to ENSO, QBO, and stratospheric aerosol forcing shows a
similar pattern over the tropical areas in the two periods (Fig. 4.5). A stronger
negative response to the ENSO forcing exists over the tropical eastern Pacific; this
indicates a negative contribution to the stratospheric warming constrained over the
same areas (Fig. 4.1b, d, e). The QBO forcing (Fig. 4.5c, d) always produces a
negative anomaly over the northern high latitudes and a positive anomaly over the
tropical areas in the two periods. In addition, stratospheric aerosols contribute
negatively to the opposing anomalies over both polar areas in the two periods
(Fig. 4.5e, f). In contrast, the tropospheric response to ENSO forcing shows an
opposite distribution to the stratosphere over the tropical eastern Pacific (Fig. 4.6a,
b). The wavelike pattern of temperature trend (Fig. 4.2) over the middle-high
latitudes is connected jointly by the ENSO, QBO, and stratospheric aerosol forcing
(Fig. 4.6). However, the wave structure shows significant differences in the two
periods. The QBO contribution to tropospheric temperature was reduced dramatically (Fig. 4.6c, d), while stratospheric aerosols largely made a negative contribution to the temperature in the troposphere (Fig. 4.6e, f).
60
A.M. Powell Jr. and J. Xu
Antarctic was partially dominated by positive trends in the period of 1978–2002,
while the opposite was generally true in the period of 1958–1977 with the Arctic
partially dominated by positive trends and the Antarctic completely dominated by
negative trends.
For the troposphere, a pronounced positive response to ENSO is observed over
the tropical eastern Pacific (Fig. 4.6a, b), which is opposite to its counterpart in the
stratosphere. This positive ENSO response in the tropical troposphere is present in
the two study periods, while the wavelike response occurs in both middle-high
latitudes. But the structure of the wavelike response shows a significant difference
in the two periods. It implies a different impact of the ENSO events on the
temperature variability over the middle-high latitudes in the two periods.
Compared to the stratosphere, the tropospheric response to QBO was significantly reduced (Fig. 4.6c, d), especially in the tropical areas. The wavelike response
is again identified over the middle-high latitudes although the area exceeding the
statistical significance test at the 95% level is markedly compressed. In addition, the
different responses in the two periods are observed over both middle-high latitudes.
Finally, the tropospheric response to the stratospheric aerosols shows (Fig. 4.6e, f)
an opposing pattern to the one in the tropics (Fig. 4.5e, f). The stronger response
appears over both middle-high latitudes. The amplitude of the response in the period
of 1958–1977 is much higher than its counterpart in the period of 1978–2002. Note
that the significant negative contribution from stratospheric aerosols, which are
impacted via volcanic eruptions, was identified over most areas.
To summarize, the linear trend of the lower stratospheric temperature (Fig. 4.1)
and the strongest temperature anomalies occur repeatedly over the Arctic, but its
sign is positive in 1979–2002 and negative in 1958–1978. The opposite response to
solar cycle forcing (Fig. 4.3) is only observed over the northern middle-high
latitudes. The response to ENSO, QBO, and stratospheric aerosol forcing shows a
similar pattern over the tropical areas in the two periods (Fig. 4.5). A stronger
negative response to the ENSO forcing exists over the tropical eastern Pacific; this
indicates a negative contribution to the stratospheric warming constrained over the
same areas (Fig. 4.1b, d, e). The QBO forcing (Fig. 4.5c, d) always produces a
negative anomaly over the northern high latitudes and a positive anomaly over the
tropical areas in the two periods. In addition, stratospheric aerosols contribute
negatively to the opposing anomalies over both polar areas in the two periods
(Fig. 4.5e, f). In contrast, the tropospheric response to ENSO forcing shows an
opposite distribution to the stratosphere over the tropical eastern Pacific (Fig. 4.6a,
b). The wavelike pattern of temperature trend (Fig. 4.2) over the middle-high
latitudes is connected jointly by the ENSO, QBO, and stratospheric aerosol forcing
(Fig. 4.6). However, the wave structure shows significant differences in the two
periods. The QBO contribution to tropospheric temperature was reduced dramatically (Fig. 4.6c, d), while stratospheric aerosols largely made a negative contribution to the temperature in the troposphere (Fig. 4.6e, f).
60
A.M. Powell Jr. and J. Xu