Thursday, May 26, 2016

What Is the PDO?

Anyone who follows seasonal predictions, including those critical to Wasatch Weather Weenies like snowfall, is immediately confronted with an acronym stew of climate indices such as ENSO, PNA, AO, NAO, PDO, IPO, QDO, MJO, etc.

This post is concerned with the PDO, or Pacific Decadal Oscillation, because I receive many questions and comments regarding its use for seasonal snowfall outlooks over western North America.  

What is the PDO?

If we skip over the gory statistical details, the PDO is simply the dominant pattern of sea-surface temperature (SST) variability in the North Pacific.  The positive (a.k.a. warm) phase of the PDO features a horseshoe of relatively warm (compared to climatology) SSTs along the west coast of North America and the eastern subtropical North Pacific with relatively cool SSTs over the midlatitude western and central Pacific.  
PDO Positive Phase.  Source: Wikipedia.
During the "cool" or negative phase, the pattern is reversed.


By the mid 1990s, scientists recognized that an abrupt, widespread change in the climate of the North Pacific Basin occurred in the late 1970s.  They were also identifying shifts in ecological indicators, such as production levels of Pacific salmon, that occurred on time scales of about a decade.

Steven Hare, a fisheries scientist at the University of Washington, first used the term Pacific Decadal Oscillation in 1996.  Subsequently, Nate Mantua, an atmospheric scientist at the University of Washington and an avid fisherman, led the seminal paper identifying the main characteristics of the PDO and its relationship to salmon production in 1997.  

Mantua et al. (1997)
Nate's contributions in this area are evidence of the Anthony Doerr quote that "there is a connection between thinking and fishing mostly because you spend a lot of time up to your waist in water without a whole lot to keep your mind busy."  

Driving mechanisms

The PDO was identified statistically and is not a single phenomenon.  Research over the past 25 years shows that the PDO reflects the combined influence of several atmospheric and oceanic processes operating at differing time scales.  One is ENSO, the El Nino/Southern Oscillation, with El Nino and La Nina reflecting the warm and cold phases of ENSO, respectively.  The relationship between the PDO and ENSO is apparent in the illustration above, which in addition to showing the North Pacific SST pattern associated with the PDO positive phase, also shows a tongue of anomalously warm water in the tropical Pacific consistent with El Nino, the warm phase of ENSO.  ENSO contributes to the PDO by influencing the mid-latitude atmospheric circulation over the North Pacific.  Other mechanisms affecting the PDO include interactions between the Aleutian Low and the North Pacific Ocean, a phenomenon known as "reemergence" in which ocean temperature anomalies from the prior cool season persist at depth in the ocean and reemerge the next cool season, and other aspects of ocean dynamics in the north Pacific.  Yeah, there's a lot going on, as illustrated by the "summary" view below.
Source: Newman et al. (2016)
The key point is that the fluctuations in North Pacific SST that we call the PDO are a reflection of a several processes operating in the atmosphere and ocean.  

Implications for climate variability and prediction

The fact that the PDO is a reflection of many processes has a number of implications for climate variability and prediction.  For one, correlation with the PDO does not necessarily imply causality.  For example, a relationship between the PDO and snowfall in a given region could reflect an alternate forcing mechanism, such as ENSO, rather than the PDO.  Assigning causal linkages to the PDO should be done very cautiously.  For two, combining the PDO with other modes of climate variability like ENSO to try and improve predictions for any given season partly represents a form of "double counting" since the two are intrinsically linked.  Whether or not this ultimately degrades the usefulness of seasonal predictions is unclear, but one needs to be cautious.   Like an onion, the more you peel it, the more it stinks.  

For those who are really interested in gory details about the PDO, a review article by Matt Newman and colleagues entitled The Pacific Decadal Oscillation, Revisited, is about to appear in the Journal of Climate.  It's paywalled, but freely available in early release mode to those of you from campus or who have a subscription for American Meteorological Society journals.

Correction: The Pacific Decadal Oscillation, Revisited, is public access, so anyone can have a look.  Thanks to the authors (or their funding agencies!) for making that happen.  


  1. Cool stuff! Is the physical cause of ENSO well understood?

    1. I think the basic mechanics are understood.

  2. Hi Jim--Nice writeup; more succinct than me. One comment: the paper is published under Open Access, so it won't be behind a paywall when it comes out.

    1. Matt:

      Thanks for the info on the open access. I always forget that the AMS journals have that option now (if paid by the authors). That's great news.