SVD analysis of SLP vs SST fields
from Kaplan et al. [2000]

Results are described in this paper
(see Section 5 and Figures 11 and 12):

Kaplan A., Y. Kushnir, M.A. Cane, 2000: Reduced space optimal interpolation of historical marine sea level pressure: 1854-1992, J. Climate, 13, 2987-3002. To download in PDF

In brief:
Using the reduced space marine sea level pressure (SLP) analysis introduced in the paper above, and the previously done sea surface temperatures (SST) analysis [Kaplan et al., 1998; PDF], we seeked to identify the leading patterns of global climate variability during the common period of these two analysis products. We applied the running 5-yr mean filters to the April-to-March annual averages of these data sets and calculated the covariance matrix between the two fields. We then performed a singular value decomposition (SVD) analysis of this covariance matrix to determine the orthonormal sets of spatial patterns for SLP and SST fields that maximize temporal covariance between their projections on these fields. The first three SVD modes were interpreted as (1) the long-term trend, (2) Pacific Decadal Oscillation (PDO), and (3) North Atlanic Oscillation (NAO). These time series were shown in Fig. 12 of Kaplan et al [2000] and are made accessible below.

Time series of the three leading modes (1856-1990) from this SVD analysis between SLP and SST are accessible here through the clone of the Ingrid Data Library.

If you only need the time series for the mode interpreted as the PDO (2nd leading mode), direct links to them are given on this page.

[Please note: Access above to the Ingrid-based Data Library clone is via port 81 of our data server. Some firewalls disallow access to any ports other than the default one (port 80). If this is the case, you will see these links hanging forever upon the click, or will get an error message. Possible solutions are to change your firewall settings to allow the access to port 81, or access the links from the outside of your firewall. If these approaches are impossible, please write to me, I can provide data directly]

Last update: December 3, 2011

Acknowledgements: This work was supported by NOAA Grant UCSIO - 10775411D / NA47GPO - 188 to the LDEO of Columbia University.
Ingrid and Ingrid-based Data Library webserver software were developed by Dr. M. Benno Blumenthal (IRI of Columbia University).