PDO forcing in observations and model

1
PDO forcing in observations and model South
Pacific SST variations and forcing and
atmospheric feedback
2
(No Transcript)
3
Reconstruct SST from AR-1 process and forcing
indices, evaluate leading EOF of reconstruction
of SST, compare with coupled model output.
Schneider and Cornuelle 2005
4
Schneider and Cornuelle 2005
5
(No Transcript)
6
Anomalous Heat Budget
7
Ocean Impact on Sea Surface Temperature
anomalous advection
8
31
SST 160E-160W 5S-5N ann rms 0.8K
SLP EOF1 ann rms 4.5mb
9
Kuroshio-Oyashio Extension Forcing and Feedback
10
Decadal Variations of SST in the South Pacific
PCM 150 years of control run Austral
winter-centered annual averages Jan-Dec high
pass lt 50 years period
11
South Pacific EOFs of SST, very inefficient
representation
EOF 1 2 3 4 5 6 ...
14 13 9 8 4 3
12
EOF of South Pacific sea level pressure /mb
Correlation of EOF 1 2
.... with EOF of S.Hemisphere -0.88
0.87 with NINO3.4 0.37
0.24
Frequency spectra of PC are white.
13
Lack of match between EOFs of SST and 1 SLP

14
SST reconstruction from heat budget
anomalous advection
magnitude of terms are smaller than North
Pacific.
15
Thermocline processes dominate the low frequency
variance
16
High frequency meridional advection anomalies
consistent with SLP
17
Thermocline memory
Thermocline depth off New Zealand
Lagged correlation of thermocline processes with
SLP Suggestive of wave propagation
Thermocline off New Zealand leading SLP no
correlation, no detectable feedback to sea level
pressure
18
Conclusions

• The North Pacific Decadal Oscillations results
  from the superposition of the SST response to
  changes in the Aleutian Low (see also Davis 1976)
  and oceanic adjustments in the Kuroshio-Oyashio
  Extension.
• The PDO is the dominant mode because the
  atmospheric sea level pressure anomalies are
  spatially organized, whilst random in time.
• The SST variations in both hemispheres are
  roughly consistent with AR-1 physics forced
  directly by the atmosphere, or through a delayed
  response due to ocean dynamics.
• In the South Pacific of the model, no EOF of SST
  clearly dominates - there is no Southern
  Hemisphere counterpart to the PDO.
• This reflects the lack of a spatial organization
  of the SLP pattern. In the Southern Hemisphere,
  SLP appears random in time and in space.

19
PAU
20
EOF of South Pacific SLP/mb
21
(No Transcript)
22
(No Transcript)
23
PCM Processes in the Kuroshio-Oyashio Extension
Observations
PCM
24
(No Transcript)
25
PCM Feedbacks to the Kuroshio-Oyashio Extension
Correlation KOE ocean sea level leads atmosphere
by one year SLP perturbation of order of 1 mb
26
Control
27

• The PDO of the coupled model is dynamically
  similar to observations.
• Differences reflect shortcomings of the PCM mean
  state.
• In the Kuroshio-Oyashio Extension, PCM attributes
  changes of SST largely to thermocline depth
  anomalies, rather than to zonal advection.
• PCM shows a consistent, local atmospheric
  response to ocean induced changes in the
  Kuroshio-Oyashio Extension.
• The feedback does not project on the excitation
  SLP pattern of the variations.
• In the model, the PDO is not a climate mode.

28
(No Transcript)
29
The Impact of Re-emergence
30
(No Transcript)
31
(No Transcript)
32
(No Transcript)
33
PCM Forcing of the Kuroshio-Oyashio Extension
34
(No Transcript)
35
(No Transcript)
36

• The PDO results from a superposition of forcing
  by ENSO, NPI (interpreted as intrinsic
  mid-latitude variability), and zonal advection
  anomalies in the Kuroshio Extension PDEL.
• The forcing footprints are non-orthogonal and
  determine the PDO. The PDO is not a climate mode.
• The contributions of the forcing are frequency
  dependent. At yearly time scales, NPI dominates,
  at interannual time scales anomalies NPI and ENSO
  are on par, at decadal time scale, NPI, ENSO and
  PDEL are of equal importance.
• The impact of the PDO is a reflection of shared
  forcing.

37
(No Transcript)
38
(No Transcript)
39
Robustness of PDO Reconstruction The Tropical
Index
Deser et al. 2004 CTI not reliable tropical
index. OTI captures low frequency changes of NPI
better. How about the impact of OTI and PDO Side
point NPI is key variable, also seen by
orthogonalization.
40
(No Transcript)
41
'Impact' of the PDO
42
Hypotheses for Pacific decadal variability
43
(No Transcript)
44
Atmosphere Ocean GCMs
45
STD of SST (contours) and temperature at 200 m
simulated by OGCM (gt0.3C/0.6C N/S of 20N) forced
by anomlous wind stress only, SST and SSS are
relaxed to climatology. Note that only special
areas (KOE in winter) act as a window of the
surface layer to thermocline anomalies.
46
(No Transcript)
47
To understand the Pacific Decadal Oscillation we
need to understand the processes affecting the
leading empirical orthogonal function of SST in
the Pacific north of 20N Approach Reconstruct
SST in the North Pacific from indices of
intrinsic variability of the Aleutian Low El
Nino adjustment of the ocean circulation in the
Kuroshio/Oyashio compare leading empirical
orthogonal functions of SST reconstruction and
observations Data NCEP/NCAR reanalysis sea
surface temperature, sea level pressure and wind
stress focus on July to June averages
48
July to June annual averages (Newman et al. 2003)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
If ??(x) ??0 separable
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
(No Transcript)
62
(No Transcript)
63
(No Transcript)
64
(No Transcript)
65
(No Transcript)
66
Hypotheses for decadal variability of the
Pacific ocean integrates atmospheric
noise Oceanic wave processes communicate
anomalies to remote areas such as the western
boundary regions
67
(No Transcript)
68
Hypotheses for decadal variability of the North
Pacific
Ocean mixed layer
69
Hypotheses for decadal variability of the North
Pacific
70
Hypotheses for decadal variability of the North
Pacific
Ocean mixed layer
Ocean mixed layer
71
(No Transcript)
72
(No Transcript)
73
(No Transcript)
74
(No Transcript)
75
What is the Pacific Decadal Oscillation? The
leading empirical orthogonal function of SST in
the Pacific north of 20N
76
Correlation 0.53 0.57
0.74
77
(No Transcript)
78
(No Transcript)
79
(No Transcript)
80
(No Transcript)
81
(No Transcript)
82
(No Transcript)
83
(No Transcript)
84
(No Transcript)
85
Tn ??n-1 ?i Fi, n
86
Observed
87
Observed
88
Observed
89
(No Transcript)
90
(No Transcript)
91
(No Transcript)
92
(No Transcript)
93
(No Transcript)
94
(No Transcript)
95
(No Transcript)
96
0.5
97
(No Transcript)
98
(No Transcript)
99
Interactive sea surface temperature increases the
variance in the atmosphere (Barsugli and Battisti
1998). Sea surface temperatures affect the
atmosphere, but effect is weak, sensitive to
background state, and involves changes of the
storm track (e.g. Peng and Whitaker 1999).
Differences of warm (1968 to 1972) and cool (1982
to 1986) epochs