Geoscience Reference
In-Depth Information
North
China
rainfall
5
7
1
Nile
floods
6
India
droughts
2
El
Nino
Java
tree rings
3
4
Figure 23.7. Statistically significant correlations (significance shown in brackets
below) between China rainfall, Indian droughts, Java tree rings, Nile flood height
and El Ni no occurrences in Peru for different time intervals between 1740 and 1984.
(1)isElNino and northern China rainfall 1770-1879 (0.08); (2) is El Ni no and
Nile floods 1770-1879 (0.08) and 1880-1984 (0.001); (3) is El Ni no and India
droughts 1770-1869 (0.03); (4) is El Ni no and Java tree rings 1770-1984 (0.01);
(5) is Nile floods and northern China rainfall 1870-1984 (0.01); (6) is Nile floods
and India drought 1770-1887 (0.05) and 1880-1984 (0.01); (7) is India droughts
and northern China rainfall 1870-1984 (0.001). Nile floods and Java tree rings
correlated significantly during the 1750s and from 1870 to 1929 (0.01). (Adapted
from Whetton et al.,
1992
, fig. 25, with supplementary data from Whetton and
Rutherfurd,
1994
.)
annual pressure anomalies associated with the Southern Oscillation extend well bey-
ond the two tropics in both hemispheres. However, the correlations between different
localities, although statistically significant, do not account for all of the year- to-year
variation in the strength of the summer monsoon, but they do account for a modestly
important part of the interannual rainfall variability. Finally, at periodic intervals,
for reasons we still do not understand, the spatial pattern of variation changes quite
abruptly, so that two localities that were previously in phase suddenly cease to be so.
A corollary to this is that future changes in the links between floods, droughts and
ENSO events are to be expected and may offer surprises. Both of these conclusions
accord with the views expressed throughout the three latest IPCC reports (IPCC,
2007a
; IPCC,
2007b
; IPCC,
2007c
).
