RealClimate debunks one myth (about global cooling) while advancing another (that warming is linear)

The climate science deniers favorite myths are about cooling. They have cooling myths about the past — see “Killing the myth of the 1970s global cooling scientific consensus.” They have cooling myths about the present (asserting that we’ve been cooling since 1998) — see “Very warm 2008 makes this the hottest decade in recorded history by far.” And they claim that recent studies predict future cooling.

In regards the last claim, a favorite new study is “Has the climate recently shifted?” (Swanson and Tsonis, 2009). I have previously noted how absurd it is for deniers to cite that study — see New study quoted by Cato Institute deniers concludes “warming over the 21st century may well be larger than that predicted by the current generation of models.”

Now RealClimate has published a long post by Swanson, “Warming, interrupted: Much ado about natural variability.” Swanson says that we’re NOT cooling, that he never predicted we will start cooling, but that we are in for about a decade of not much warming.

That view requires believing 1) something very unusual happened in the past decade and 2) global warming is linear. Now the second assumption myth has no basis in climate science, a conclusion so obvious that I’m certain RealClimate agrees. Before addressing that myth, here is what Swanson says on the first point:

Everything hinges on the idea that something extraordinary happened to the climate system in response to the 1997/98 super-El Ni±o event (an idea that has its roots in the wavelet analysis by Park and Mann (2000). The figure [above] shows the spatial mean temperature over all grid boxes in the HadCRUT3 data set that have continuous monthly coverage over the 1901–2008 period. While this provides a skewed view of the global mean, as it is heavily weighted toward North America, Europe and coastal areas, unlike the global mean temperature it has the cardinal virtue of being a consistent record with respect to time. The sole exclusion in the figure is the line connecting the 1997 and 1998 temperatures.

It may well be that something unusual did happen, since something unusual seems to have happened in upper ocean warming:

Remember that figure is how Pielke Sr. justifies saying, “upper ocean warming has halted since 2003,” when one could say just as easily — and more accurately from a climatology perspective, which looks at longer term trends — “upper ocean warming has soared since 2002” (see “Roger Pielke Sr. also doesn’t understand the science of global warming “” or just chooses to willfully misrepresent it”)

But Swanson’s conclusion that we will see another decade of flat temperatures is based on a very erroneous assumption. Swanson writes:

What we find is that when interannual modes of variability in the climate system have what I’ll refer to as an “episode,” shifts in the multi-decadal global mean temperature trend appear to occur….

The contentious part of our paper is that the climate system appears to have had another “episode” around the turn of the 21st century, coinciding with the much discussed “halt” in global warming. Whether or not such a halt has really occurred is of course controversial (it appears quite marked in the HadCRUT3 data, less so in GISTEMP); only time will tell if it’s real. Regardless, it’s important to note that we are not talking about global cooling, just a pause in warming.

What’s our perspective on how the climate will behave in the near future? The HadCRUT3 global mean temperature [below] shows the post-1980 warming, along with the “plateau” in global mean temperature post-1998. Also shown is a linear trend using temperatures over the period 1979–1997 (no cherry picking here; pick any trend that doesn’t include the period 1998–2008). 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 Ni±o. 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. Of course, this contrasts sharply with other forecasts of the climate system; the purple line roughly indicates the model-based forecast of Smith et al. (2007), suggesting with a warming of roughly 0.3 deg C over the 2005–2015 period.

Before getting to the nonsense of linear warming the above figure depicts, let me first note that while Swanson uses the Hadley Center’s data, the NASA Goddard Institute for Space Studies data is almost certainly superior (see “What exactly is polar amplification and why does it matter?”). Remember, “there are no permanent weather stations in the Arctic Ocean, the place on Earth that has been warming fastest,” as New Scientist explained (see here and here). “The UK’s Hadley Centre record simply excludes this area, whereas the NASA version assumes its surface temperature is the same as that of the nearest land-based stations.” Thus contrary to what the global warming deniers say about the recent temperature record, it is almost certainly the case that the planet has warmed up more this decade than NASA says, and especially more than the UK’s Hadley Center says.

So that’s why I see the NASA temperature record as more accurate, which puts 2005 as the warmest year on record, with a rough tie for second between 2007 and 1998. And of course that kills about half of Swanson’s theory, since it means there is little evidence we hit a plateau 10 years ago.

But what really kills his theory is the straight-line warming trend he draws from 1950 to 2030 of about 0.1°C per decade. That has no basis in climate science. As atmospheric concentrations rise, the rate of warming increases. That, of course, is how Hadley itself (!) gets total warming of 5.5°C or more by 2100, the vast majority of which occurs this century (see “Hadley Center: “Catastrophic” 5–7°C warming by 2100 on current emissions path”).

Here, for instance, is a Hadley scenario for 980 ppm and 5.5°C warming from preindustrial levels from a 2003 Geophysical Research Letters (subs. req’d) paper, “Strong carbon cycle feedbacks in a climate model with interactive CO2 and sulphate aerosols.”

Not exactly linear from 1950 to 2030 and beyond.

It is just plain odd for Swanson to base his analysis on Hadley temperature data, but then ignore Hadley’s own analysis of how temperature is likely to rise this century on the business as usual emissions path.

In fact, “Smith et al, 2007” which Swanson plots in his figure is a Hadley (!) paper from Science: “Improved Surface Temperature Prediction for the Coming Decade from a Global Climate Model” (see “Climate Forecast: Hot “” and then Very Hot”), which concluded:

Our system predicts that internal variability will partially offset the anthropogenic global warming signal for the next few years. However, climate will continue to warm, with at least half of the years after 2009 predicted to exceed the warmest year currently on record.

I’m liking that prediction more and more (see NOAA says “El Ni±o arrives; Expected to Persist through Winter 2009–10”³ “” and that means record temperatures are coming and this will be the hottest decade on record). Here’s what Hadley thinks will happen:

So, cheers to RealClimate for letting Swanson kill the myth that his paper predicts global cooling, but jeers for letting Swanson push the notion of linear warming.

And I’ll be happy to take a bet with anybody on RealClimate (or with Swanson) that Hadley is right that “at least half of the years [in the decade] after 2009 will exceed the warmest year currently on record.”

UPDATE: Tamino has a good post on this here:

I have two overriding opinions of this work. My first overall opinion is that I don’t believe it’s correct, for several reasons…. Also, the hypothesis uses certain aspects of the HadCRU temperature time series which aren’t shared by the GISS or NCDC time series, so at least in part it’s dependent on the use of a particular data set.

The paper also uses linear trend rates over 7-year intervals to suggest that the most recent lower trend rate (the suggested “plateau”) may indeed match the criteria for a “state change” of the climate, because it doesn’t coincide with a known cause (la Ni±a or a volcanic explosion)…..

I’m especially skeptical of the suggestion that we may be beginning an extended “plateau” of temperature change. Using GISS data, the observed variations in trend rate since 1975 (on 7-year or longer time scales) really are statistically indistinguishable from random noise, and the residuals from a straight line fit 1975-present are likewise indistinguishable from noise. Of this I’m sure: although it’s certainly possible that the climate has recently “shifted,” as yet there’s no sound statistical evidence to confirm it (nor does the paper, or the RealClimate post, claim such evidence)….

Furthermore, the trend on which they base their projection is from 1979 to 1997 (centered on 1988), which is one of the lowest 18-year trend rates in the recent GISS record. Both earlier and later time spans of the same length indicate greater warming, in particular the time span 1974 to 1992 gives a linear regression trend rate over 0.02 deg.C/yr….

Tamino provides a good figure on this.

Bottom Line: Swanson would appear to have cherry picked a very low trend rate — from 1979 to 1997 — and assumed that is a linear trend that will continue through 2020 (to 2030), in the third figure above (that I reposted from RealClimate). That is the only way he can come to the conclusion that we will see little or no warming over the next decade. That also requires ignoring the off-trend-line data since 1997 and explaining it away as an anomaly, which, again, there is little basis for doing. Finally, it requires ignoring other data (like NASA GISS), but then at the same time ignoring Hadley’s analysis of projected temps over the next two decades.

Whether Tamino is right that Swanson’s paper is important, only time will tell. But the RealClimate post is analytically very weak and based on an incredibly dubious set of assumptions.

Related Posts: