Posted by: Carl | April 17, 2009

A Short Demonstration of the Long-Term Effects of ENSO

In an old post, I explored the underlying temperature signal in each ocean by removing the El Nino/Southern Oscillation (ENSO).  I then compared the underlying signal to ENSO to explore any long-term effects of particular El Nino events.  The events of 86/7 and 97/8 appeared to result in strong, long-term increases in temperature of the Indian and Pacific Oceans.  I also speculated about a potential role for ENSO in North Atlantic SST.  In this short post, I will improve my past description of of the long-term effects of ENSO on SST.

Since the global climate shift of 1976, SST variation in the Indian and West Pacific Oceans can be characterized entirely by step-change responses to those two El Nino events and the eruption of Mt. Pinatubo in 1991.  This is demonstrated below.

sst-in-west-pacific-and-indian-oceans3As informative as this graph is, the data is averaged over a large portion of the Earth’s ocean.  The entire region represented by the graph does not necessarily represent every peice of ocean within the outlined area.  So to get a better idea of exactly what regions exhibited step change responses to the two El Ninos, I’ve pointwise correlated the time series shown in the graph above with global SST during two periods, surrounding the 86/7  and 97/8 El Ninos.  The correlations were run during periods surrounding the event, so that the step change behavior of the time series is emphasized.  Slight alterations in the period used for the correlations does not significantly alter the results.

The correlation map surrounding the 86/7 El Nino (using data from the years 1984-1990) is shown below.


The correlation map surrounding the 97/8 El Nino (using data from the years 1995-2001) is shown below.


The regions in red represent those that experienced step changes up in temperature during the El Nino events.  SST anomaly data from these regions do confirm the suggested behavior.  It is clear that the Indian ocean, as well as parts of the West Pacific, responded to the two El Nino events with long-term step changes.  For context, below is a correlation map of Nino 3.4 (an index describing ENSO) and SST.


At least concerning the 97/98 El Nino, it appears that those regions in the Pacific that experienced step changes in SST are also the regions that are negatively correlated with ENSO.

North Atlantic SST have also been driven by ENSO, through the modulation of the Atlantic Meridional Overturning Circulation (AMOC), the Atlantic branch of the Thermohaline Circulation.  Since the global climate shift of 1976, modeled AMOC data from the European Center for Medium-Range Weather Forecasts shows a strong negative correlation between the flow rate of the AMOC and ENSO.  This is shown in the graph below, with the Cold Tongue Index used to describe ENSO.


If the AMOC data is correct, it appears that the 1997/8 El Nino was immediately followed by a long-term slowing of the AMOC flow rate.  Although the causes of the Atlantic Multidecadal Oscillation (the AMO is defined as trend-removed North Atlantic SST) have not been entirely pinned down, the AMOC is suspected to play a role.  In the graph below, I’ve compared the running total (which could be thought of as an integral) of the AMOC data with the AMO.  I am using the running total of the AMOC flow rate because the temperature of the Atlantic at a given point in time is not just dependent on the speed of the AMOC at the time, for its initial condition is dependent on previous variation in the flow rate.  Below is a graph comparing the two.

amoc-vs-amo1The correlation is fairly strong, and the general shape of both curves are very similar.  After the immediate effect of the 97/8 El Nino, the North Atlantic maintained very warm SSTs, possibly due to a slowed AMOC associated with the El Nino.  However, this analysis suggests that a slowing of the AMOC will raise temperatures, and some would surely object this argument.  I might investigate the relationship between the AMO and the AMOC further in a future post.  However, even if the AMOC has not driven AMO recently (which evidence suggests is not the case), the fact that the AMO never returned to normal levels after the 97/98 El Nino implicates the “El Nino of the Century” in the recent variation of North Atlantic SSTs.

This post clarifies my previous post, attributing recent SST variation in the North Atlantic Ocean, Indian Ocean, and parts of the Pacific Ocean to the 1986/7 and 1997/8 El Nino events.  In future posts, I will investigate the causes of the observed step changes.  It appears that ENSO’s influence on cloud cover in the Pacific and Indian oceans may have driven the observed step changes.


Sea Surface Temperature; NCDC v3 ERSST;

AMOC Flow Rate; European Centre for Medium-Range Weather Forecasts;

Cold Tongue Index; Joint Institute for the Study of the Atmosphere and Ocean;

Nino 3.4; NCDC v3 ERSST;



  1. […] us anything about the source of warming.  I’ve outlined all of this in previous posts (1, 2, 3) , so there’s no reason to copy graphs […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: