Tropical Teleconnections and Western Boundary Current Variability

1
Tropical Teleconnections and Western Boundary
Current Variability

• Michael Alexander
• NOAA/Earth System Research Lab
• WBC Workshop January 2009

2
WBC Web page (http//www.cdc.noaa.gov/WBC)
3
(No Transcript)
4
El Niño La Niña Composite
DJF SLP Contour (1 mb) FMA SST (shaded ºC)
5
(No Transcript)
6
Bridge related question 1

• What impact does the wind stress (curl)
  associated with ENSO have on the Kuroshio via
  Rossby wave generation?
• If impact is modest (Schneider), why is this the
  case since it so strongly projects on PDO
  forcing? Frequency of forcing?
• Role of ENSO in maintaining NPGO related wind
  stress forcing.

7
Correlation PDO SAT
Obs
8
Correlation between ENSO and the PDO
Why do IPCC-class models under estimate the
connection between the PDO and the tropical
Pacific (ENSO)?
9
Composite Niño - Niña SST (C shaded)
Yr 0 Nino 87, 91, 97, 02, 06 Nina 88, 98, 99, 05
ASO (yr 0)
Mean Con Int 4C
10
Bridge-related question 2

• What is impact of ENSO on WBCs in seasons other
  than winter?
• e.g. Kuroshio in summer
• Are there feedbacks on the atmosphere?
• If so does it involve different processes than in
  winter?
• While Atlantic signal has been known before high
  resolution SST data reveal its tightly confined
  to GS/North Atlantic Current.

11
Storm tracktrajectory counts of surface lows
Climatological Mean JFM
Niño -Niña
Alexander et al. 2006 J Climate
12
Composite El Niño - La Niña JFM
Alexander Scott 2008 J Climate
13
Bridge Related Questions 3

• Course resolution AGCM-MLM suggest a positive
  feedback of bridge-related SSTs in the Atlantic
  on the NAO
• Is this robust?
• Role for intense air-sea interactions over GS?
• Non-linearity of ENSO response (not shown) much
  stronger response during La Niña then El Niño.
• If robust, why?
• Issues for studies that linearly remove ENSO
  before conducting statistical analyses.

14
Atlantic teleconnections
Peng et al. 2005, 2006
15
Observed Lead-Lag MCA SLP SST

• Model studies suggest that tropical Atlantic
  impacts N. Atlantic in winter (with extratropical
  air-sea feedback) results from observations much
  less clear.

Frankignoul and Kestenare 2005
16
Indian Ocean Teleconnections
Cor w GOGA Pacific
Z 500
Cor w GOGA Atlantic
Precip
Hoerling et al. 2004 Climate Dynamics
17
Indian Ocean - NAO teleconnection

• How robust is this connection especially with
  poor match for circulation over the Pacific.
• Model dependence - precipitation over the Indian
  ocean much strong in CCM (Hoerling et al) than in
  CAM (Deser and Phillips 2006).
• Indian Ocean steadily warming (most likely due to
  global warming) - so getting this signal right
  could be very important.
• If this teleconnection is realistic is there
  feedback from the N. Atlantic Ocean.

18
Tropics ltgt Extratropics
19
SFM Model Experiment - Additional Heat Flux
Forcing (Wm-2) NDJFM(0)
SST (C)
20
Experiment - Control
Winds SST MJJ(0)
21
Di Lorenzo et al.
22
Tropical - Extratropical Interactions
Observation
NP
TP
Deser et al. 2004
23
Life Cycle of NPM
SST
HT400
Regression on KOE SST
-16 yr
-12 yr
-8 yr
-4 yr
0 yr
4 yr
24
Origin of NPM Subtropics vs. Subpolar
Modeling Surgery Experiment
Origin Subploar Route !
25
Extratropics gt Tropics

• What processes are important?
• Atmospheric (e.g Kwons talk)
• Direct response to SSTs?
• WES
• Shift of ITCZ to the warm hemisphere
• Several could be involved with SFM
• Ocean
• Rossby waves gt Kelvin waves on western boundary
• Subduction
• Appears to be overturning rate rather than
  advection of anomalies
• How important relative to tropical only
  variability?

26
A Relay Teleconnection
Return Atmospheric bridge
N. Pacific
(WES)
EQ
STC
Positive
Tropics
Extratropics
Feedback
Figure 11 Schematic diagram of the
extratropical-tropical relay teleconnection. A
persistent strengthening ( weakening) of the
Aleutian Low creates cooling (warming) in the
western-central North Pacific and warming
(cooling) in the eastern subtropical Pacific the
latter rapidly propagates to the tropics through
the WES feedback to create warming ( cooling) in
the equatorial region nearly synchronously this
fast surface coupled process operates annually
and creates persistent westerly (easterly)
anomalies in the subtropics, leading to a
weakening ( strengthening) of the STC to further
sustain the warming ( cooling) in the tropics.
As a result, the warming (cooling) in the tropics
can further intensify ( reduce) the Aleutian Low
through return atmospheric bridge, forming a
positive feedback loop.
Wu et al. (2007, Climate Dynamics)
27
Issues/Directions

• Getting tropical-extra tropical connections (e.g.
  ENSO-PDO) is critical
• As key as any local WBC metric for WBC
  variability
• New datasets that suggest teleconnections can be
  WBC focused (e.g. ENSO gt Gulf Stream).
• What causes this.
• Non-normal/nonlinear impact of remote
  teleconnections
• Sampling, are they robust?
• What causes nonlinearities
• Methods to remove ENSO
• Model experiments to isolate forcing and
  feedbacks
• (w/wo) air-sea interaction
• Model surgery
• Consider a set of model experiments across models

28
(No Transcript)
29
Extratropical gt Tropical Connections
Seasonal Footprinting Mechanism (SFM)
Subduction
Meridional cross section through the central
Pacific
(SFM Vimont et al. 2003 Subduction Schneider
et al. 1999 JPO)
30
Di Lorenzo et al.
31
Subduction and the Subtropical Cell
Transport at 9ºN 9ºS
Convergence SST
McPhaden and Zhang 2002 Nature
32
Climate Indices
1900
2000
(Boreal Winter)
25
47
77
- NP Index
33
Experiment Design

• Model
• AGCM CCM3
• Reduced Gravity Ocean (Cane-Zebiak) Model 30S-30N
  in Pacific.
• Slab model over remainder of the ocean
• Models are anomaly coupled
• 100-year Control run
• SFM Experiment
• Add additional heat flux forcing associated with
  the NPO
• 20S-60N similar results when forcing gt 10N
• Initiate 60 heat flux anomaly runs from Nov in
  control run.
• Apply Heat flux anomaly during first Nov-Mar
• Then let model evolve with unperturbed fluxes for
  12 more months.
• Compare ENSO evolution in perturbation and
  control runs.
• Note model already includes SFM

34
Seasonal Footprinting Mechanism
35
Additional SFM Forcing

• NPO from AGCM
• With Climatological SST
• Isolates intrinsic variability
• 2nd EOF of SLP EOF in North Pacific in Winter
• Regress Sfc Heat flux on PC
• double flux values
• Max values of 30 Wm-2
• Add identical/constant forcing in each of the
  experiments

36
(No Transcript)
37
Deser and Phillips 2006
38
Bader and Latif 2005
39
The Atmospheric Bridge
Meridional cross section through the central
Pacific
(Alexander 1992 Lau and Nath 1996 Alexander et
al. 2002 all J. Climate)
40
FIG. 10. The El Niño La Niña composite of U200
(m/s) during JFM(1) for (a) EKM, and (b) ?. The
shading (contour) interval in (a) is 1.5 (3) m/s.
In panel (b), the shading indicates 95 and 99
confidence limits for the ? U200 (contour
interval 0.5 m/s).
41
FIG. 9. The El Niño La Niña composite of Z500
(m) during JFM(1) for (a) observed (1950-1999),
(b) EKM, (c) MLM, and (d) ?. The shading
(contour) interval is 5 (10) m in (a)-(c). In
panel (d), the shading indicates 95 and 99
confidence limits for the ? Z500 (contour
interval 5 m).
42
FIG. 3. The El Niño La Niña composite of SST
(K) during JFM(1) for (a) observed (1950-1999),
(b) EKM, (c) MLM, and (d) EKM-MLM (?). The
shading (contour) interval is 0.1 (0.5) K in
(a)-(c). In panel (d), the shading indicates 95
and 99 confidence limits for the ? SST (contour
interval 0.1 K). The large box (panels (b)-(d))
indicates the region of prescribed SST forcing
and the smaller boxes (panel (d)) in the N.
Pacific and N. Atlantic will be used for regional
averaging (see FIG. 8).
43
Experiment - Control
Thermocline depth (m)
44
Subduction
Central North Pacific
Colored contours -0.3C anomaly isotherms for 3
different pentads
Black lines mean isopycnal surfaces (lines of
constant density)
Averaged over 170ºW-145ºW
45
Do subducting anomalies reach the equator and
influence ENSO?
a)
b)
c)
d)
Latitude
Year
46
Mechanism for Atmospheric Circulation Changes due
to El Nino/Southern Oscillation
Atmospheric wave forced by tropical
heating
Latent heat release in thunderstorms
Horel and Wallace, Mon. Wea Rev. 1981
47
Mechanism Loop of NPM
SP SST
Salinity gt SP Conv. Fbk
Oyashio
Ocean or Coupled Mode ?
KOE SST
Atmos. Resp
S. Curl gt M-lat. Thm Fbk
Sfc wind -- gt WES Fbk
N. Curl -- gt SP Thm Fbk --
SP Rossby Wave Delay Strong enough to break conv
48
ENSO NOAA High Res Uwnd composite10S-60 N
across the N. Pacific and N. Atlantic
49
Niño - NiñaObserved JF(1) Wind Stress/SST
50
Propagation SSH (Color)
SSH_S
SSH_T
50N
Westward Propagation
45N
40N
35N
Westward signal clear in subpolar Subpolar SSH
dominated by salinity
x
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
Change in Subduction Rate
Transport at 9ºN 9ºS
Convergence SST