Connecting ENSO, PDV, and the North and South Pacific

Connecting ENSO, PDV, and the North and South Pacific

http://wattsupwiththat.com/

A new paper in Geophysical Research Letters was brought to my attention by Dr. Leif  Svalgaard.  

Tropical origins of North and South Pacific decadal variability by Jeremy D. Shakun and Jeffrey Shaman makes some very interesting findings suggesting that both the northern and southern Pacific Ocean has evidence of the Pacific Decadal Variation PDV being driven by ENSO variations. They produced a model, which when run correlates reasonably well with observations.

Fig 4. Observed (red line with circles) and modeled (blue line with slashes) PC1s for the (top) North and (bottom) South Pacific. The model is of an AR-1 process forced by ENSO, see the paper for details - click for a larger image

 

Abstract:

The origin of the Pacific Decadal Oscillation (PDO), the leading mode of sea surface temperature variability for the North Pacific, is a matter of considerable debate. One paradigm views the PDO as an independent mode centered in the North Pacific, while another regards it as a largely reddened response to El Nin˜o-Southern Oscillation (ENSO) forcing from the tropics. We calculate the Southern Hemisphere equivalent of the PDO index based on the leading mode of sea surface temperature variability for the South Pacific and find that it adequately explains the spatial structure of the PDO in the North Pacific. A first-order autoregressive model forced by ENSO is used to reproduce the observed PDO indices in the North and South Pacific. These results highlight the strong similarity in Pacific decadal variability on either side of the equator and suggest it may best be viewed as a reddened response to ENSO.

They write about the graph above:

…we model PDV as a first-order autoregressive process driven by ENSO as done by Newman et al. [2003]. This AR-1 model is applied to the North and South Pacific separately.

The modeled PDO index at year n is a function of the modeled PDO index at n – 1 and the observed ENSO index (Nino 3.4) at n. These annually-averaged indices are centered on boreal winter (Jul–Jun) for the North Pacific and austral winter (Jan–Dec) for the South Pacific. Per Newman et al. [2003], the coefficients β and α are parameters derived, respectively, by regression of the PDO index on the ENSO index, then autoregression of the residual time series with a lag of one year. h is an uncorrelated noise term not used in our analysis but shown for completeness. a and b are 0.51 and 0.56 for North Pacific PC1 and 0.62 and 0.71 for South Pacific PC1. While Newman et al. [2003] found this simple model did a remarkable job reproducing the observed 20th century PDO index in the North Pacific (r = 0.63 in our study), it yields an even stronger fit to our Southern Hemisphere PDO index (r = 0.71) (Figure 4).

The greater success of the model in the South Pacific may be a function of its larger α and β terms, which indicate that the persistence of SST anomalies and ENSO forcing are more important. The stronger ENSO signal in the South Pacific may derive from the equatorial asymmetry of ENSO SST anomalies in the eastern tropical Pacific, which extend considerably farther to the south than to the north. One implication of this finding is that the South Pacific may be a better place to develop paleo-ENSO records as it appears to contain a ‘cleaner’ ENSO signal.

Conclusion

Deriving a Southern Hemisphere equivalent of the PDO index shows that the spatial signature of the PDO can be well explained by the leading mode of SST variability for the South Pacific. Thus, PDV appears to be a basin-wide phenomenon most likely driven from the tropics. Moreover, while it was already known PDV north of the equator could be adequately modeled as a reddened response to ENSO, our results indicate this is true to an even greater extent in the South Pacific.

Leif has a copy of the paper on his website that you can read here

Citation:

Shakun, J. D., and J. Shaman (2009), Tropical origins of North and South Pacific decadal variability, Geophys. Res. Lett., 36, L19711, doi:10.1029/2009GL040313.