2011-10-13 12:47:43Continued Lower Atmosphere Warming
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Rebuttal to Pielke's "no TLT warming since 1998" myth.

Continued Lower Atmosphere Warming

2011-10-13 13:03:22Averaging period?
Glenn Tamblyn

glenn@thefoodgallery.com...
60.228.59.135

Just a check on your diagrams  Dana.  They still look rather jagged for a 5 year running average. This what I get from GISTemp for a 60 month average. Are they 60 month or maybe a smaller period by accident?

 

 

2011-10-13 13:41:10
Tom Curtis

t.r.curtis@gmail...
112.213.163.226

How can you have a five year running average ending in 2011 in 2011?

 

Further, Pielke in a recent comment questioned not including all data up to Sept, 2011.  Clearly he is intent on including the 2010/11 La Nina in the data.  Although this requires more work as you then need to discuss the La Nina as well, it avoids any accusation by him of cherry picking.  (Because of the issue with the graphs, I'm not sure if you have inlcuded the September data or not.)

 

Finally, I've stolen your thunder a bit with my post here:

http://skepticalscience.com/news.php?n=1025&p=2#65252

 

If Pielke responds to that, that will give you some pointers as to how he will respond to your blog, so it's worthwhile keeping an eye out for that.

2011-10-13 13:47:07
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Whoops brain fart, it's a 12-month running average, not 5-year.  I included data through the latest month available, which is actually August 2011 (at least on Wood for Trees where I got it from).

2011-10-13 13:54:47
Rob Painting
Rob
paintingskeri@vodafone.co...
118.92.35.63

So Dana, is this going to be on hold until we've finished our exchanges with him, or what?

2011-10-13 14:09:44
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

I don't think so Rob.  I toned it way down so as not to offend Pielke, but I see it as being in the same vein as the 'CO2 contribution to warming' post.  We gave Pielke the opportunity to admit his error, but he stood behind it, so this post is the response, explaining the error.

2011-10-13 19:28:30
Dikran Marsupial
Gavin Cawley
gcc@cmp.uea.ac...
88.108.235.105

Looks good to me.  Two points:  Firstly picking the start date to get a flat trend is just as much cherry picking as shoosing to start in an anomalously warm year.  It is hard to think of a better definition of cherry picking that making a choice to make an argument as strongly as possible when you know that a slightly different choice refutes the argument, and that is exactly what Pielke has openly admitted to doing!  Secondly regarding the non-linearity issue, I specifically challenged Pielke to provide evidence that the signal is significantly non-linear over these sorts of timescales and he ducked the question, so you could give a link to my question and his (feeble) answer.

If Pielke does respond to this post, I'll engage him on hypthesis testing there so as not to disrupt the discussion of point #3 on the other thread.  I think it would be polite to send Pielke an email alerting him to the new post. 

2011-10-13 22:08:17
Tom Curtis

t.r.curtis@gmail...
112.213.161.10

Looks good to me as well.

2011-10-14 02:12:52
Kevin C

cowtan@ysbl.york.ac...
144.32.196.109

Very good indeed.

I like Tamino's analysis a lot, but I'm not sure the final plot is up to peer-review standard: I think it overestimates the effect of the solar cycle owing to the short period (~3 cycles) and coincidence with the lagged effect two volcanoes and an increase in aerosols. The actual size of the solar variation using his own two-box energy balance model gives a solar contribution of 1/3 to 1/4 the size he gets here. I've posted a comment about this on Tamino's article, but he has only responded tangentially.

I doubt Pielke will look deeply enough to pick up that issue, but he might complain it's not peer-reviewed work.

2011-10-14 02:48:30
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Thanks guys.  Since there aren't any major comments, I might post this today so that we can move the TLT discussion to this thread, and focus on OHC in the Disagreements thread.

Dikran - I certainly agree that regardless of his reason, Pielke's 1998 choice was a cherrypick.  But I specifically scrubbed that term from the post to keep it as civil as possible so as not to offend his delicate sensibilities :-)  I'll add links to your comments on the non-linearity.

Kevin - true, but Pielke has referenced non-peer-reviewed work too (i.e. Lucia's blog), so he can hardly complain.  I think the graph is just a good illustration.

2011-10-14 04:25:04
Dikran Marsupial
Gavin Cawley
gcc@cmp.uea.ac...
139.222.14.107

experimenting with woodfortrees, I came up with this plot

 

plotting the data without the 1998 el-nino spike makes it clear how much the apparent cooling depends on the presence of the El-Nino spike, without it there seems little reason to thing the slope in the green bit of the plot is any different from the slope in the red bit.

2011-10-14 05:26:48
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

Dikran, I calculated the slopes for the two windows 1979-1997 (inlcusive) (The major El Nino started in the spring of 1997), and 1999-2011.  The post 1998 trend is in fact steeper than that beforehand!

1979- end of 1996: slope +0.034 C/decade 

1979-end of 1997:  slope 0.041 C/decade

1999-end of 2010: slope 0.21 C/decade

1999-present: slope 0.18 C/decade.

But then he will just say that we are cherry picking ;)

2011-10-14 05:37:35
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

Dana,

I found this from Tamino in December 2010:

"None of the 10-year trends is “statistically significant” but that’s only because the uncertainties are so large — 10 years isn’t long enough to determine the warming trend with sufficient precision. Note that for each data set, the full-sample (about 30 years) trend is within the confidence interval of the 10-year trend — so there’s no evidence, from any of the data sets, that the trend over the last decade is different from the modern global warming trend."

 

2011-10-14 05:40:02
Dikran Marsupial
Gavin Cawley
gcc@cmp.uea.ac...
139.222.14.107

Indeed. The 1985-current slope looks pretty constant, so I think the difference is largely down to the 1979-1985 bit.  A Pielke is using visual test, this gives a visual test that shows the problem with his visual test ;o)

2011-10-14 05:54:35
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

dana:

"However, part of the reason the TLT data is "~flat" over that period is that 1998 was an anomalously hot year."

i would say "warm" rather than "hot".

Not sure if this is relevant, but it is new, a paper by Mears et al. (2011).  look at their Figure 12 and Table 1.

2011-10-14 06:03:26
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

Last post Dana ;)  It looks good to me, firm but not uncivil or rude.

Maybe consider also adding a trend line from 1999-present in Figure 3, just to emphasize the huge difference that one year makes when dealing with a) such a huge outlier as a start point, and b) such a short window of time.

I would also consider adding my proposed question to Pielke on the other thread to end off with trying to sum up his position on calculating trends:

"Pielke appears to be of the opinion that there is nothing wrong statistically with picking a clear outliers as starting and end points for calculating a short-term trend <15 years), and then suggesting that the resulting trend is robust/meaningful".

Or something like that.

[Update] Dana, Pielke is now trying to move the goal posts by looking at trends from 2002 (an even shorter window!), so maybe nip that one in your post.

2011-10-14 06:14:04
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

Oh forgot this.

2011-10-14 06:35:05
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

I updated the post with Pielke's comments about looking at trends since 2002 (seriously?  He doesn't seem to be getting the message about short-term noise and statistics at all) and reorganized the end a bit.  I may throw in a trend since 1999 later, as Alby suggests.

2011-10-15 01:13:57
thingsbreak

things.break@gmail...
66.7.151.194

I was thinking about leaving a comment like this (after popping in some citations):

 

I find it incredibly strange that Dr. Pielke believes that brief periods of "flat" temperatures- even in the absence of large eruptions- are supposed to challenge the big picture of long-lived GHGs exerting an increasingly dominant forcing on the climate system.

He's well-aware at this point that such behavior is an emergent property within individual modeling runs of even CMIP3 era models.

I agree that the present period is not without interest. Whether or not there is actually "missing heat" in obs (per Trenberth) or a combination of unrealistically efficient ocean mixing in models and larger aerosol cooling (per Hansen) are certainly worth investigating.

However neither (or for that matter anything Pielke himself has contributed) particularly change the long-term expectations of the climate response to unchecked emissions growth in a way that justifies claims (implicit or otherwise) that "global warming has 'stopped'".

Perhaps Dr. Pielke is not aware of how strange such claims (be they about surface/lower trop warming trends, or say Arctic sea ice decline) sound to general and academic audiences alike. In my experience, most people don't interpret statements like his to reflect variability about a long term trend (which is, after all, the mainstream interpretation that Pielke is seeming objecting to).

2011-10-15 01:51:22
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

I'm OK with that TB.  He is now talking about the sun, and a whole bunch of other bogus arguments:

"It amazes me that with the diversity of human climate forcings, the newly recognized higher importance of solar forcing,"

Actually, what is truly amazing is how far some people will go to defend a lie Dr. Pielke....

He providers no citations so it is not obvious what he is talking about.  And him citing Hansen is counter to his argument, not for it.

 

2011-10-15 01:55:16
Dikran Marsupial
Gavin Cawley
gcc@cmp.uea.ac...
139.222.14.107

His reply to my post asking him if the difference in the pre-2002 - post 2002 trends was statistically significant was amusing.  He said "What are they?"

BTW, I have nearly sorted out an estimate of the power of the test, doesn't look too clever at the moment.  I'll post it when I am sure I have it right.

2011-10-15 02:18:57
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

Hi Dikran,

Maybe we should start a new thread in the General chat area to continue this.  

Thanks for your work on this.

2011-10-15 02:25:41
Dikran Marsupial
Gavin Cawley
gcc@cmp.uea.ac...
139.222.14.107

no problem, it was about time I learned a bit more about time series analysis ;o)

we could set up a new chat thread, but it looks as if Pielke has copped out

"Lets agree that it is a short term flattening in the lower tropospheric temperatures (since 2002) that we will all follow to see if it persists in the coming months and years. This short term trend does not tell us that the long term trend has been significantly changed (yet)."

2011-10-15 02:58:54
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Pielke also admitted the reason he's made this argument is to undermine the usefulness of the TLT record, to get us to focus on the OHC record.  Ironically, he made a similar argument about OHC too (no warming since 2003).

I gave him a bit of a telling off for giving the delayers ammunition just to further his goal of getting people to focus on OHC.

2011-10-15 03:26:34
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

Dana et al.,

Please see my comments on the new thread at "General Chat"-- I would suggest moving all comments there.  I hope that there is some useful information there for you, a lot of ranting by me, but hopefully also some useful information.  I think we need to start getting tougher/firmer with him.

2011-10-15 03:50:35
thingsbreak

things.break@gmail...
66.7.151.194

Dana,

 

I think it was a HUGE mistake to bring up the whole bit about providing cover for inactivists. Please stop doing that with Pielke.

1. It has nothing to do with science, skeptical or otherwise. It's politics.

2. Pielke will use it as an excuse to avoid previous questions.

3. Those previous questions/the science-only stuff are more than damning on their own.

4. Pielke is on record saying we should do something (even if he gives tons of ammo to inactivists) so he has tremendous plausible deniability.

5. It's hugely counterproductive to the ability to use Pielke's previous comments about being pro-mitigation as a cudgel in other contexts if he's being tarred as anit-mitigation here (implicitly, by proxy, etc.).

2011-10-15 03:57:04
Albatross
Julian Brimelow
stomatalaperture@gmail...
199.126.232.206

ThingsBreak,

I don't know he is accusing we warmists of misleading policy makers.  Do you disagree that him using the 1998 start date as evidence that AGW stopped is not providing ammo for the deniers and "skeptics"?  

My two cents.

2011-10-15 15:08:14
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

I very much diagree, tb.  There are two major problems with Pielke's arguments

1) Scientific/statistical inaccuracy

2) He's providing fodder for fake skeptics, whether intentionally or not.

Pielke has said we need to reduce CO2 emissions, yet he's aiding those who are trying to prevent that goal.  He doesn't seem to realize he's undermining his own goal.  This needs to be pointed out.

2011-10-15 18:31:25Good Cop
Glenn Tamblyn

glenn@thefoodgallery.com...
121.216.112.192

I just put up an intentionally rather waffly philosophical post on this thread, trying to play a bit of Good Cop/Bad Cop. See if we get any reaction from Pielke (or others) about the moral responsibilities associated with putting up posts that can be misconstrued by the eager beavers of the D-Machine. Maybe nothing will come of it, maybe he might respond at a philosophical level. This might (possibly remote chance) be a means of prying open his psychological can-of-worms.

You guys are doing well. Keep it up (that strange noise you can hear is applause from the dugout). No matter what you do or don't achieve in terms of 'turning' him, for the Denialosphere to get much traction from this will require jumping through gazillions of hoops.

Wild card hope. Turning Pielke Snr into a quazi-ally. The more we can 'cut' the moderates out from the herd of the hard core D'ers the better. Keep stroking his vanity. Anyone who can put up a photo of the 'old guy with the grey beard looking deeply thoughtful' is full of it.

2011-10-15 23:25:59
nealjking

nealjking@gmail...
84.151.41.220

From my point of view: The discussion seems to have gone well:

- Our side made focused, relevant points

- It seemed to me that it was reasonably clear that Pielke was persistently refusing to admit a severe problem in his analysis

- Although many people pointed this out, I did not get the impression of a "piling on". The tone was mostly reasonable.

 

Overall, I think we did ourselves some good in this exercise. Two recommendations:

- The discussions by Glenn and Dikran et al. should be expanded into an article on statistical tests, the null hypothesis and what it means to disprove it. This is a very important pillar of understanding. It deserves a full article or three.

- Someone (JC?) should get an outside view of how things looked: Do we have any friends at RC or wherever who could give an objective view?

 

I think this was a great effort. I think the discipline exercised in NOT jumping in all the time was very beneficial to the overall effect. I still think we need to change the format for debates (Guest vs. Designated SkS Debater, plus peanut-gallery comments on the side). But the effort at discipline helped.

 

2011-10-16 03:22:02
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Yeah I think it went well.  Pielke came out of every discussion looking really bad.  First royally screwing up the CO2 contribution calculation, then instead of admitting he was wrong, coming up with new wrong excuses why his original argument must somehow still be right.  Very unscientific.

Then in this one he essentially admits that his argument was a cherrypick (after spending a lot of time unsuccessfully trying to defend it), but criticizes us for doing real scientific and statistical analysis instead of pandering to cherrypicking fake skeptics.

Also each time we catch him in these errors, he claims it's on an unimportant subject, even though he's devoted blog posts, presentations, etc. to each point.  Really, Pielke couldn't have come out of this looking much worse.  I'm sure he'll write a blog post trying to spin the discussion to make himself look good, and the fake skeptics will run with that, but to any unbiased observer, we pretty much whooped his butt.

In the end we come off as the scientifically strong side, and Pielke comes off as clearly very biased, and consistently advocating an unscientific approach.  Not bad for a blog vs. a climate scientist.

A post on statistical significance might be useful, elaborating on the fundamental wrongess of the Pielke-type approach.