Sheldon Walker gets into a pickle picking pauses at WUWT

Sheldon Walker has written another two articles for Anthony Watts' blog WUWT (archived as Part 1 and Part 2). What Sheldon was trying to do was prove there was a Pause (with a capital P) in global surface warming recently. He didn't manage that, though he thinks he did. What I'm going to show is that in order to pick a Pause, Sheldon had to:

• carefully select a 13-year period for a linear trend (aka cherry-pick)
• avoid an interval just two years longer
• ignore a finding he started out with, that even 15 years is too short to be assured of a meaningful trend
• ignore another finding that non-overlapping trends suggest real differences
• accept that a Pause is not the same as a stop!

Pretty plots

In part 1, he posted some pretty pictures, which are 16,920 plots of trends going from January 1975 to December 1999.  Each trend line covers at least 10 years. He gave examples to illustrate:

• 10 year trends:  e.g. January 1975 to January 1985, February 1976 to February 1986 etc
• 10 year and 1 month trends:  e.g. February 1980 to March 1990
• 10 year and 2 month trends:  e.g. January 1985 to March 1995
• 20 year trends:  e.g. September 1979 to September 1999

As Sheldon said, "This might appear to be overwhelming, but with Excel and a modern computer, it can be calculated quite easily."

Here is one of his charts:

Figure 1 | Trends from 1975 to 1999 for periods from 10 years to 25 years. Credit: Sheldon Walker at WUWT

A valuable lesson about plotting trends for short periods - they are not robust

Sheldon wrote what should be a valuable lesson for him and WUWT-ers (my dot points and emphasis):

This graph holds a lot of valuable information, but it needs a little interpretation.

For example, how does the warming rate change with the trend length.
From the graph:

• The warming rate for 10 year trends varies from -0.20 to +2.80 degC/century
• The warming rate for 15 year trends varies from +0.65 to +2.20 degC/century
• The warming rate for 20 year trends varies from +1.02 to +1.61 degC/century
• The warming rate for 24 year and 11 month trends doesn’t vary at all, because there is only one, which is for the entire period, and it equals +1.71 degC/century

These results probably agree quite well with most people’s expectations. One lesson is, be wary of 10 years trends. You can get just about any warming rate that you want from a 10 year trend. Note than in certain circumstances a 10 year trend can be meaningful, but in general, 10 years trends are all over the place.

Indeed. Pick a short period such as ten years and you could find almost any trend you want, from a negative trend to a huge trend. 15 year trends vary hugely too, which should give one pause (I couldn't resist writing that). Sheldon didn't say under what circumstances a 10 year trend could be meaningful. He seems to have tossed that in just in case he wanted to use a 10 year trend somewhere.

Distinguishing periods

Sheldon goes further and plots two different time periods on the same chart.

Figure 2 | Trends from 1950 to 1974 and from 1975 to 1999 for periods from 10 years to 25 years. Credit: Sheldon Walker at WUWT

The green trends down the bottom are for the period 1950 to 1974, the top yellow-orange is 1975 to 1999. Sheldon astutely noticed that the trends in the earlier 25 year period were lower (y or vertical axis) once you take trend lines of around 15 years or longer (x or horizontal axis at the bottom). Sheldon wrote:

Note how there is no overlap between the 2 graphs for trend lengths greater than about 15 years. This reinforces the idea that these 2 time intervals have very different warming rate profiles.

So far so good. That was a fun way to illustrate how there are different temperature trends depending on:

• the length of the period
• the start and end points.

Sheldon's big question and bad science

Then Sheldon wrote something rather strange, which has no basis in science. He said:

Now, the BIG question. If you add together these 2 periods, [1950 to 1974] and [1975 to 1999], and calculate the warming rate for the combined interval [1950 to 1999], what would the warming rate be? I have done this, and a linear regression over the combined interval has a warming rate of +1.12 degC/century. OK, but what does this value of +1.12 degC/century actually represent.

It is NOT the warming rate for normal anthropogenic global warming.

It is NOT the warming rate for when there is NO anthropogenic global warming.

It is an artificial average rate of warming, for an interval when anthropogenic global warming was present for about 1/2 the time, and absent for about 1/2 the time.

I've no idea where he got his notions about what caused global warming in different periods. He doesn't say. He just jumped right in there and made some unsubstantiated (and wrong) pronouncements. Anthropogenic global warming has been with us for a very long time. Some scientists have provided evidence that it's been with us since humans started altering the landscape, way back when. We've definitely been affecting climate since we started adding copious amounts of greenhouse gases to the air above us, and we started that in earnest in the 19th century.

Figure 3 | Merged annual CO2 record from 1722 to 2015. Atmospheric CO2 record based on ice core data before 1958, (Ethridge et. al., 1996; MacFarling Meure et al., 2006) and yearly averages of direct observations from Mauna Loa and the South Pole after and including 1958 (from Scripps CO2 Program)Data source: Scripps CO2 Program

The AR5 IPCC report states (section TS 4.8):

There is strong evidence that excludes solar forcing, volcanoes, and internal variability as the strongest drivers of warming since 1950. 

And in the Summary for Policymakers, the best estimate is that all of the warming in the period since 1951 is from human activity:

The best estimate of the human-induced contribution to warming is similar to the observed warming over this period.

More specifically in TS.4.2:

Greenhouse gases contributed a global mean surface warming likely to be between 0.5°C and 1.3°C over the period between 1951 and 2010, with the contributions from other anthropogenic forcings likely to be between –0.6°C and 0.1°C and from natural forcings likely to be between –0.1°C and 0.1°C.

What Sheldon means by "artificial rate" he doesn't explain either. Maybe he'll come here and help me out.

How Tom Karl confuses Sheldon

Sheldon then jumped into the paper by Karl et al (2015), which I've written about previously. He was very confused (and wrong again) when he wrote:

Unfortunately, Karl et al used this value as their “normal” anthropogenic warming rate, and based on this value, they concluded that the warming rate for [2000 to 2014] did NOT support the notion of a global warming “hiatus”.

Recapping quickly on the Karl et al paper:

Karl et al adjusted the NOAA data to account for the 0.12 degC average difference between buoy and ship SSTs. This “correction” had an impact on temperature trends, with the largest impact being on trends from 2000 to 2014 (which is where “The Pause” was meant to be).

Sheldon is a bit mixed up. In Karl15 the authors state that the biggest impact on the trend from 2000 to 2014 wasn't the ship-buoy difference, it was the corrections to ship data from the shift from buckets to engine intake thermometers:

Of the 11 improvements in ERSST version 4 (13), the continuation of the ship correction had the largest impact on trends for the 2000-2014 time period, accounting for 0.030°C of the 0.064°C trend difference with version 3b. (The buoy offset correction contributed 0.014°C dec⁻¹ to the difference, and the additional weight given to the buoys because of their greater accuracy contributed 0.012°C dec⁻¹ ...) . 

Anyway, Sheldon doesn't dispute the need for the corrections. His complaint was that the authors of Karl15 used the same periods for comparison as the IPCC did in its section in AR5 on the so-called "hiatus". He thinks it should have been a different comparison. He explains:

Now [1975 to 1999] is an interval having significant anthropogenic global warming.

But [1950 to 1974] is an interval having very little anthropogenic global warming.

By joining these 2 intervals together to form [1950 to 1999], Karl et al have created an interval that basically has half strength anthropogenic global warming (half with warming, and half without warming). But Karl et al used this value as their “normal” anthropogenic warming rate, when they compared it to [2000 to 2014].

I'm not fussed about that. The IPCC section on the so-called "hiatus" used similar periods, which was why they were chosen by the authors of Karl15. You could argue that they were both wrong to do so, but that's not the problem other people had. The biggest thing was that a ten or fifteen period is too short for any comparison, as Sheldon's charts show. In an article at realclimate.org, Stefan Rahmstorf said the same thing, writing: "trends over such short periods are not particularly robust".

If you thought that Sheldon was about to point this out, you'd be disappointed. Instead of going back to his earlier conclusions, which were:

One lesson is, be wary of 10 years trends. You can get just about any warming rate that you want from a 10 year trend. ... in general, 10 years trends are all over the place.

That applies to his 15 year trends as well. As you saw, for the period from 1975 to 1999, the trend for 15 year segments varied from +0.65 to +2.20 degC/century. Instead of pointing out that a short trend doesn't have much meaning, Sheldon wrote:

There are 2 simple ways to explain how [2000 to 2014] could have half strength anthropogenic global warming.

1) The period [2000 to 2014] could consist of 2 parts, one part which has anthropogenic global warming, and one part which does NOT have anthropogenic global warming (like [1950 to 1999]). But I do not think that this is the case.

2) The more reasonable explanation is that the period [2000 to 2014] has a lower warming rate than “normal” anthropogenic global warming. The warming rate would be about 50% of the “normal” warming rate. This could be called a “Slowdown”, a “Hiatus”, or a “Pause”. Whichever name you prefer, the data shows that it exists.

Which only goes to show that some people are remarkably inconsistent. Why did Sheldon bother to show how short term trends are all over the place? It spoils his argument. You could say his earlier workings don't just spoil his argument, they destroy it.

His argument is also destroyed by Cahill15, who found that there was no change in the trend from the 1970s to 2014.

Figure 4 | Overlaid on the raw data are the mean curves predicted by the three CP model. The grey time intervals display the total range of the 95% confidence limits for each CP. The average rates of rise per decade for the three latter periods are 0.13 ± 0.04 °C, −0.03 ± 0.04 °C and 0.17 ± 0.03 °C for HadCRUT, 0.14 ± 0.03 °C, −0.01 ± 0.04 °C and 0.15 ± 0.02 °C for NOAA, 0.15 ± 0.05 °C, −0.03 ± 0.04 °C and 0.18 ± 0.03 °C for Cowtan and Way and 0.14 ± 0.04 °C, −0.01 ± 0.04 °C and 0.16 ± 0.02 °C for GISTEMP. Source: Cahill15

Trying to prove a Pause

In Part 2, Sheldon said how he wanted to prove the existence of a Pause in global warming, with a capital P. He wrote:

After my first article, I decided to use MTA [multi-trend analysis] to try and prove that the Pause exists. I wanted to compare a graph of the interval where the Pause existed, against a graph of an interval where “The Pause” did not exist (a reference interval). If there was a significant difference, and it was the right kind of difference (e.g. a lower warming rate), then I would have good evidence that the Pause existed.

Sheldon explained that it was very important to get the start and end years right (aka cherry-pick) or the Pause with a capital P would disappear. He wrote:

Picking the right intervals was important. From my previous investigations into the Pause, I knew that the core years were from 2002 to 2013. This is a 12 year interval with a very low warming rate. Moving the start date to 2001 increased the warming rate slightly, but gave a longer slightly weaker Pause. Moving the start date to 2000 increased the warming rate even more, but gave an even longer weaker Pause. Moving the finish date to 2014 also increased the warming rate, and moving the finish date to 2015 weakened “The Pause” considerably, because of the 2015 El Nino.

 Sheldon settled on periods from 8 to 13 years, even though he previously stated how these periods gave wildly diverging rates. Here is his final chart, with trend rates from 8 to 13 years for three periods:

• January 1975 to December 1987
• January 1988 to December 2000
• January 2001 to December 2013

Figure 5 | Trends for three different periods, with trends plotted from 8 to 13 years. Credit: Sheldon Walker at WUWT

As you can see, except for periods from 10 to 13 years, the trend charts overlap. What that means is that there were some 8 year segments in all three periods that had similar trends. Same for 9 and ten year segments. It's only with the longer periods that you see the difference.

Remember how Sheldon was pleased to differentiate the periods

Now as you recall, in Part 1 Sheldon wrote how a lack of overlap reinforces that time intervals had "very different warming rate profiles". Well, there's precious little time interval that has no overlap in Figure 5 above.  The longest is not much more than two years. So using Sheldon's own criteria, the periods are not differentiated.

Sheldon's Pause didn't stop the warming

There is probably one other thing that you noticed: Sheldon's Pause with a capital P didn't stop the warming. There was no Pause at all. Almost all the area covered by his blue trend line is positive, above zero, which means it continued to warm over almost all of his umpteen trend lines from January 2001 to December 2013. Now if you look at the area in yellow-orange, which is the period from January 1975 to December 1987, you'll see that more of those trend lines were in more negative territory, hitting as low as minus 1 C/century.

More signs that the Pause is imaginary

What you might not have noticed is that if you plot the trend lines you'll find some interesting things. To set the scene, here are Sheldon's three intervals again, with NOAA data:

Figure 6 | Global mean surface temperature change from 1880 to 2015. The different coloured sections show the periods that Sheldon Walker compared as shown in the legend at the bottom of the chart. Data source: NOAA

Next here is a chart of global mean surface temperature from Sheldon's start year of 1975, with a linear trend line at 1.66 C/century. Last year the temperature anomaly was quite a bit above the trend:

Figure 7 | Global mean surface temperature change from 1975 to 2015 with linear trend. Data source: NOAA

Now here's a comparison of the trends of all the 13 year periods Sheldon chose to plot:

Figure 8 | Global mean surface temperature change from 1975 to 2015. Linear trend are shown for 13-year periods marked. Data source: NOAA

Here are the linear trends for each of those 13 year periods:

• January 1975 to December 1987 is 2.09 C/century
• January 1988 to December 2000 is 1.37 C/century
• January 2001 to December 2013 is 0.52 C/century.

What happens if we pick another period ending sometime this century, say in 2005, which is five years into Sheldon's Pause. Well, blow me down. Will you look at that! What a difference it makes.

Figure 9 | Global mean surface temperature change from 1975 to 2015. Linear trends are shown for 13-year periods marked. Data source: NOAA

Here are the linear trends for each of those 13 year periods:

• January 1975 to December 1987 is 2.09 C/century
• January 1988 to December 2000 is 1.37 C/century
• January 1993 to December 2005 is 2.58 C/century.

It certainly doesn't seem to be a very solid Pause if by shifting the 13 year window back a little bit results in a steeper slope than either of the non-Pause periods. Here are some more numbers:

• 1994 to 2006 is 2.18 C/century
• 1995 to 2007 is 1.82 C/century
• 1996 to 2008 is 1.51 C/century

In fact ending more than 60% into the pause, in 2008, which means a start year just before the first year of the 97/98 super-El Nino, is the first time we see the trend drop below the long term trend from 1975 to 2015 (1.66 C/century). 

Sheldon's Pause loses its rank at 15 years

Next is what happens to Sheldon's Pause if he'd chosen a 15 year period:

Figure 10 | Global mean surface temperature change from 1975 to 2015. Linear trends are shown for 15-year periods marked. Data source: NOAA

It looks as if we've found another Pause. This time it's a slower Pause than the one that ends in 2013. The trends are:

• January 1980 to December 1994 is 0.83 C/century
• January 1999 to December 2013 is 1.16 C/century.

Twenty years is another dud for Pauses, thirty years is worse

What about if we look at a twenty year stretch:

Figure 11 | Global mean surface temperature change from 1975 to 2015. Linear trends are shown for 20-year periods marked. Data source: NOAA

You won't be surprised that there's another Pause that beats the one ending in 2013. This time it's from 1977 to 2016. Here are the trends:

• January 1977 to December 1996 is 1.05 C/century
• January 1994 to December 2013 is 1.34 C/century

And if you're curious what happens when the normal climate period of thirty years is considered:

• January 1967 to December 1996 is 1.53 C/century
• January 1984 to December 2013 is 1.62 C/century

A short term slowing isn't a Pause with a capital P

The thing is that Sheldon has shown what everyone knows and no-one is denying. It's not that global warming stopped. It didn't. For a short period in the early part of this century, surface warming was at a lower rate than in some previous periods. Yet as shown by Cahill15, the difference wasn't so great as to denote a change in the trend from the 1970s.

The main reason for this is most likely the fact that the Pacific was in a cool phase. Here is a chart showing ENSO events and the PDO phases with the annual mean global surface temperature.

Figure 12 | Global mean surface temperature with ENSO years and PDO phases. Data sources: GISS NASA, Bureau of Meteorology

Sheldon thinks his Pause hasn't stopped

Sheldon adds a final word with an implicit prediction of Pausing:

A final word about the future. The Pause has been weakened by the 2015 El Nino. That does not mean that it never existed. Anybody gloating over the Pause becoming weaker, should bear in mind that El Nino’s do not last forever. Once the El Nino’s temperature increase has gone, the Pause will probably strengthen. A La Nina may also give the Pause a boost. Do not underestimate the Pause, it may surprise you yet.

El Nino's might not last forever, but atmospheric CO2 lasts a very long time. I don't think anyone will be gloating as the world continues to heat up. [Added later today at 4:18 pm:] As for Sheldon's "weakened", that's weak. That's lame. That's denier code for Sheldon losing his Pause. Extending Sheldon's cherry-picked period out two years to 2015 and the trend becomes 1.35C/century. Not a Pause in sight. BTW 2015 would have been a record hot year even without the El Nino.

Figure 13 | Global mean surface temperature change from 1975 to 2015. Linear trends are shown for extending Sheldon Walker's "Pause" out to 2015, and for the whole period from 1975 to 2015. Data source: NOAA

From the WUWT comments

Just a few, because this article is already very long.  Most WUWT deniers don't understand Karl15 still, or if they do they persist in making wrong comments. I really can't be bothered with them here. I've written ample about this. Nick Stokes had some interesting comments, like this one (too long to copy here).

richard verney quotes Sheldon, and then disputes the Pause, pointing out that the warming didn't stop:

February 25, 2016 at 8:47 am
“To be more specific, the Pause has an overall warming rate of about 27% of reference interval 1, and less than 42% of reference interval 2. These percentages represent a large reduction in the warming rate, and justify the name “Slowdown”, or “Hiatus”, or “Pause”.Could anybody deny the Pause, after seeing that evidence?"

Whilst I would accept that IF the overall warming rate has fallen to some 42% or better still some 27%, then there has been a slowdown in the rate of warming, I do not see how as a matter of the ordinary meaning of the word hiatus or pause a reduction in the rate of overall warming to those levels could properly be described as a pause or hiatus..

For there to be a pause. there would have to be a protracted period when there was no statistically significant warming at all in the time series data.
My understanding of your analysis is that is not what your analysis demonstrates.

Sheldon Walker agrees and wants WUWT-ers to start using the word "slowdown"

February 25, 2016 at 10:08 am
In my opinion, the term “Slowdown” is the most accurate name for the event that I have described.
The problem is that the terms “Pause” and “Hiatus” are the most common names used at the moment.
How do we get people to use “Slowdown”?

He's got buckley's getting the die-hard deniers at WUWT to do that. dbstealey, for example:

February 25, 2016 at 10:14 am
How do we get people to use “Slowdown”?
Hey, Sheldon, that’s a new one! ‘Slowdown’. I like your spin!
Unfortunately for the alarmist Narrative, global warming has been stopped for many years. So let’s just use ‘stopped’.
K? Thx, bye. 

Frank de Jong points out that in Sheldon's pretty plots, the middle values get much more prominence than the end points:

February 25, 2016 at 9:25 am
Sheldon,
An interesting way of looking at trends. What I am a bit worried about is that in these kinds of calculations, the points in the middle get used more often than the ones at the ends. In a way, they are weighted (much) more heavily. It then also follows that the choice of interval can influence the results by getting the “most desirable” points in the center of the interval. Maybe a sort of inverse weighting can help alleviate this problem?

Best regards,
Frank 

DanMet'al notices that Sheldon had to be very picky with his start and end dates, and more:

February 25, 2016 at 9:33 am
Sheldon
Your post title states, “Very strong graphical evidence for the Pause (Part 2)” .
According to your Part 1 post, you constructed your analysis method (to include every conceivable trend) to thwart criticism that you were cherry picking trend start dates.
As a Lukewarmer, I have two comments:
(1) You are still subject to the “cherry picking” charge because you selected a’prior the start-end dates of the three regimes in your analysis.
(2) But more importantly, your ad-hoc analysis method is mathematically incomprehensible, at least to me. Have you constructed hypothetical test cases based on a mix of trend (constant and linear) and natural variability of varying magnitudes … such that you can more clearly explicate your method and its significance.
Good to keep thinking and challenging … and I thank you for that!!
Best
Dan 

John of Cloverdale WA Australia

February 25, 2016 at 3:12 pm
One day, in the not so distant future, “The Pause” will become “The Plateau”. 

DayHay thinks an ice age is coming:

February 25, 2016 at 3:56 pm
Then the plateau will become the freaking cliff….

References and further reading

Stocker, Thomas F., ed. Climate change 2013: the physical science basis: Working Group I contribution to the Fifth assessment report of the Intergovernmental Panel on Climate Change. Cambridge University Press, 2014. (available here)

Karl, Thomas R., Anthony Arguez, Boyin Huang, Jay H. Lawrimore, James R. McMahon, Matthew J. Menne, Thomas C. Peterson, Russell S. Vose, and Huai-Min Zhang. "Possible artifacts of data biases in the recent global surface warming hiatus." Science 348, no. 6242 (2015): 1469-1472. DOI: 10.1126/science.aaa5632 (pdf here)

Debate in the noise - article by Stefan Rahmstorf at realclimate.org, about Karl15 (above)

Niamh Cahill, Stefan Rahmstorf and Andrew C Parnell. "Change points of global temperature". 2015 Environ. Res. Lett. 10 084002. doi:10.1088/1748-9326/10/8/084002 (open access)

Grant Foster and John Abraham. "Lack of evidence for a slowdown in global temperature." US CLIVAR Variations • Summer 2015 • Vol. 13, No. 3 (open access)

From the HotWhopper archives

• Sheldon Walker came to HotWhopper to get help for his WUWT article - February 2016
• Global surface warming continues without pause contrary to denier claims - February 2016
• On Seeps and SCAMS Part I: Lessons for Climate Scientists - May 2015
• NOAA: No pause in the global surface temperature - June 2015
• Global sea surface temperature and model projections, with Bob Tisdale - February 2016
• Lots of articles about the NOAA paper, Karl15
• How to go about going against mainstream science - a lesson (about the Anthropocene) for climate sceptics - April 2015