On North Atlantic Intedecadal Variability: A Stochastic View
o
, . . ,
~ . .
~
. .. ,
0.5
o
- o . •
-,
-' .0_1
_2 . :S
-,
PC · I M / RE HC ZOOm T = 1 Bo. e=9a
5"2.0
S.;JO
• • 0
:s.50
rT"Iode l ti"",,, [yeo.-s]
. ...... .
....... . .. . "
-~ · :I.*'o,-----:T:;---=:--"""", ..... 0,.--....,..;-:----:.,..,0,,-0 ---;;.,..,.o;;-----;;."'o.-----:=--~,.--~
~ode l
tim e
(yeor""s] '
165
Fig. 9.12 Time series ofthe leading POP mode of North Atlantic upper ocean heat content
(0-200m) anomalies. The POP analysis has been performed using the model years 500-700.
The dashed line shows the real part and the solid line the imaginary part. The real part time
series is highly coherent with those of the leading CCA mode (with opposite signs). From
Timmermann (1996).
a)
b)
Fig. 9.13 a) Imaginary and b) real part pattems of the leading POP mode of upper ocean
heat content anomalies. The two POP pattems describe a clockwise rotation of the heat
content anomalies. From Timmermann (1996).