Posted by: Carl | July 13, 2009

Swanson’s Not-So-Novel Post at RealClimate

Kyle Swanson, of the University of Wisconsin-Milwaukee, recently wrote a guest post at RealClimate entitled, Warming, interrupted: Much ado about natural variability.  After defense of a recent paper as having no implications for the non-existent AGW debate, Swanson writes,

“We hypothesize that the established pre-1998 trend is the true forced warming signal, and that the climate system effectively overshot this signal in response to the 1997/98 El Niño. This overshoot is in the process of radiatively dissipating, and the climate will return to its earlier defined, greenhouse gas-forced warming signal. If this hypothesis is correct, the era of consistent record-breaking global mean temperatures will not resume until roughly 2020.”

Is Swanson suggesting that it is no coincidence that the “El Nino of the century” was followed by the warmest decade of the century?

Swanson goes on to write, “Why would anyone in their right mind believe what I’ve just outlined? Everything hinges on the idea that something extraordinary happened to the climate system in response to the 1997/98 super-El Niño event (an idea that has its roots in the wavelet analysis by Park and Mann (2000)).”

Well, there’s a few of us.  And we’ve been saying it for awhile.  In fact, when I have attempted to make this exact case on alarmist websites, like RealClimate, the ad homs come flying, and I am told that by definition ENSO in a non-radiative oscillation so my ideas are silly and I am an idiot. I don’t especially care (aside from what it says about science), except it forces me to wonder how RealClimate let Swanson get away with such ridiculous claims!  Perhaps the answer lies in the fact that although we both noted the step-change resulting from the 1997/8 El Nino, he suggested it had no implications for AGW theory.  So this is how science works for the alarmists; truly, this is a great example of how they will only support ideas that support their cause.  And that’s not science.

Swanson also never quite says that the 1997/8 event was radiative, though that is what his analysis would imply.  Is that his argument?  If so, why only the 1997/8 event?  Could ENSO be a radiative oscillation?

Swanson provides the graph below to justify his claims.

"spatial mean temperature over all grid boxes in the HadCRUT3 data set that have continuous monthly coverage over the 1901-2008 period"

"spatial mean temperature over all grid boxes in the HadCRUT3 data set that have continuous monthly coverage over the 1901-2008 period"

Note the caption to that graph.  Is this really the best way to show the 1997/8 step change?  Of course not.  If Swanson was actually familiar with global SST data, as all climate scientists should be, he would know exactly what regions saw this step change; he would know what the “extraordinary” event that happened to the climate system was; he would also know that the same “extraordinary” event took place in response to the 1986/7 El Nino as well.  He is representing this step-change as a mysterious global change, when it can actually be pinned down rather precisely.  The 1986/7 and 1997/8 El Nino events caused step-changes in the Indian, NW Pacific, and South Pacific Oceans.  The 1997/8 event also caused a step-change in the North Atlantic due to long-term slowing of the Atlantic Meridional Overturning Circulation.  This does not have to be a mysterious event; the evidence is in SST data, and all you have to do is stop thinking that a time-series of global SST tells 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 here.

So now to his argument.  He suggests that the pre-1997 trend was the true antropogenic trend, and the 1997/8 step-change shot temperatures higher than the actual trend. However, this pre-1997 trend is not anthropogenic.  It is derived from the 1986/7 step-change.  SST behavior prior to 1986/7 shows no step-change or significant upward movement outside of the expected effects of ENSO (and a slight over-response to the climate shift of 1976-8).  The graph below illustrates these points.  Swanson’s background trend (assuming a step-change after the 1997/8 El Nino) is actually a product of two other step-changes, related to the 1986/7 El Nino and the 1976-8, ENSO-induced, climate shift.

ERSST v3b ENSO (scale: 1/11) subtracted from ERSST v3b global SST, lag=2 months

ERSST v3b Nino 3.4 (scale: 1/11) subtracted from ERSST v3b global SST, lag=2 months



  1. I heard it described at a conference in 1984 at Northwestern that AGW would tend to lead to a stepwise rise instead of a sawtooth climate pattern. As positive deviations would be reinforced and negative temperature deviations would be dampenede. That is what your graph reminds me of.

  2. Interesting point. However, I would point to the regional pattern of warming. The step-changes did not occur everywhere, giving us evidence that the cause is specifically related to ENSO, not random.

    In the North Atlantic, it is clearly tied to a long-term slowing of the AMOC after the 1997/8 El Nino.

    In the West Pacific/South Pacific/Indian/Indonesian throughflow, the origin of the warming can actually be seen in weekly SST maps. At the height of the El Nino, cool water begins to move Eastward along the Equator, splitting the warm El Nino water into the North & South Equatorial Currents. These currents bring the warm water Westward, where it spreads out on the surface, creating step-changes in temperature over broad-swaths of ocean.

    I have shown all of this, here:

    Bob Tisdale has a great movie showing the process, which can be found here:

  3. I wonder what Swanson’s co-author would have to say?I don’t think they agree on some of Swanson’s more baseless points.

    It’s funny, though. The research in question directly contradicts the AGW attribution argument-and yet by spinning, they can make it “consistent”.

  4. You picked up on the inconsistencies in Swanson’s post quite well. Though how else would it have got on RC if it was not mangled to be “on message”?

  5. I also identified the three steps using strucchange in R for the 1976, 1986 and 1997 breakpoints in both the HadCRUT3 and GISS datasets. It is good to see that my calculations were not isolated.

    I have postulated that the effect of each is cumulative and gives the overall signal of warming from the 1970s to present.

  6. Richard, I completely agree, though we don’t know what kind of natural variation underlies the cumulative step-changes.

    Has your analysis found any breakpoints before 1976?

  7. Carl,

    No. I did notice any other breakpoints from 1950 to the 1976 breakpoint. It is likely that there are others prior to 1950 (perhaps around 1940 and the 1920s?) but I did not investigate that.

  8. sorry. Should say “I did not determine any other breakpoints from 1950”

  9. Carl,

    You said:

    “Richard, I completely agree, though we don’t know what kind of natural variation underlies the cumulative step-changes.”

    Perhaps the release of the McLean et. al. paper (
    goes part of the way to explaining it.

  10. Richard,
    Here’s the way I see it.
    The most obvious layer of internal variation is ENSO and volcanoes. The Australian paper shows this, though it’s nothing novel. This layer has not contributed to the trend in the past few decades.
    The next layer is the long-term effects due to ENSO (1976, 1986, and 1997/8).
    However, there are longer, more gradual patterns in SST underneath this that are very difficult to find underneath the other two layers.
    The first graph in my post here is of NW Pacific SST.
    I like this graph because it shows very clearly how the step-change effects of ENSO interact with the longer, gradual variation in SST to produce the SST curve.

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: