Bishop Hill The big news down under
Jun 17, 2014 Climate: solar I hope everyone is reading the series of posts by David Evans and Jo Nova about their new hypothesis on why variations in solar irradiance apparently have such a limited effect on the planet's temperature. It's probably fair to say that many sceptics have scratched their heads on this subject from time to time, but the team from down under have gone the extra mile, coming up with what is starting to look like a fascinating explanation, namely that there is a delay between the change in irradiance and subsequent changes in temperature. They hypothesise further that this may be something to do with changes in the Sun's magnetic field.
It's too early to say whether this all holds up of course, but I'm certainly going to be keeping a close eye on it.
Reader Comments (109)
Lubos appears to have deleted his post, which was originally here: http://motls.blogspot.co.uk/2014/06/david-evans-notch-filter-theory-of.html.
David Evans has retained a copy of that post, and made it available here: http://joannenova.com.au/luboss-missing-post/.
I think it would be appropriate for Lubos to respond to David's points, and to explain the deletion.
John Shade - interesting that he deleted the original. I assume we will get a further reaction once all of David Evans material has been posted. That may also depend on how much Lubos needs to revisit his own comments.
Lubos' post, although patchy, was actually perfectly reasonable. I followed his arguments and agreed with them.
The problem with David Evan's spectral analysis is that the phase spectrum is not analysed for. I have made a comment to this effect on the lubos-missing-post thread, I made a similar comment here. Commenter Cees at joannenova, on the lubos-missing-post thread also makes a compelling argument as to why David Evan's spectral analysis is insufficent to support his argument.
As Lubos pointed out in the missing post, the "transfer function" is the spectral division of the output by the input. This is a standard analysis technique when looking at linear (convolutional processes). But it is only half of the answer. The other half is in the phase spectrum of the "transfer function". Instead of presenting the phase spectrum, David has stated it is unreliable. Well, without it his argument is founded on a tautology and cannot be supported.
The reason is partly given by Cees. If we create two random white noise sequences their amplitude spectra will each be flat (albeit noisy) lines in the frequency domain. Their spectral division will also be a flat line, but noisier.
If we add a periodic function like a sine wave with a period of 11 years to the first random series then we will get a spectrum which is flat, except there will be a spike up at 11 years. If we divide the output random spectrum we had before by this spectrum then at the 11 year frequency the output (flat level) value will be divided by a now bigger spike, and as if by magic the "transfer function" now has a spectral notch in it.
The two random series are unrelated, they are just random, but because there is a coherent signal on the first function, which is not present on the second (output) signal the "transfer function" has a notch. In other words, in the spectral domain to make the input random noise look like the output random noise we have to remove the coherent signal we added to the input. So the notch, in this context, says something on the input is missing from the output. Actually in this example its because they are physically unrelated .
However, it could be a physical process model that causes this change, for example a filter between the input and output. How do we check this? We would look for causality as one criteria. David side steps this by assuming causality and so introduces a filter delay. This is not a proof, it is a tautology.
The phase spectra would reveal more about this and resolve the issue. The division of the amplitude spectra and difference in the phase spectra form the cross-correlation. This is usually examined in the time domain and is highly diagnostic. This is the complete tool for establishing this type of relation. I use it every day in my professional job. I have downloaded the HadCRUT4 annual temp data and performed the cross-correlation with the Lean (2000) TSI data. With the TSI+Background the peak cross-correlation is quite good (at about R=0.71) but it occurs at zero lag ie no time shift or filter delay. This is bad news fof David's argument. David's argument looks to me like a tautology. And that is what Lubos argued.
"Lubos appears to have deleted his post, which was originally here: http://motls.blogspot.co.uk/2014/06/david-evans-notch-filter-theory-of.html."
More likely it's a problem with your internet connection because it's still there.
"I think it would be appropriate for Lubos to respond to David's points, and to explain the deletion."
That depends on the nature of the response. If it's a weak response, it's probably better to ignore it than begin a squabble.
"Lubos' post, although patchy, was actually perfectly reasonable. I followed his arguments and agreed with them."
I think it's waffly and he's trying to sound clever. He should be able to make his point more succinctly.
BTW Evans has replied to your comment.
There is a post script now at the bottom of the Lubos critique which should be noted
Will (7:42AM), the post was deleted for about 24 hours, and has now been reinstated. Lubos used some quite emotive language and was clearly in an impatient frame of mind when he wrote it, and this is unfortunate. Although I must confess to smiling when he gets stuck in to the 'other side' in that way. He is right to pursue vigorous criticism of papers regardless of which 'side' they come from and he demonstrates that impartiality here. But it seems there was a communications breakdown between him and Evans/Nova since they had apparently been exchanging emails on the draft paper in the months before the release. Each is accusing the other of not listening/reading carefully enough. I hope the exchanges will continue.
I realize this is late to the thread - sorry. For what it's worth, I've asked Lubos to correct the postscript and gave him permission to quote my emails direct. His response was to say he'll have to ignore my emails for the moment.
His paraphrasing bears little resemblance to my messages. You can see all those "hysterical" and "emotional" words for yourself in the emails, which I've quietly released in full. http://jonova.s3.amazonaws.com/evans/emails-to-lubos-3.pdf If he were an alarmist I would have been delighted to take apart his extra note publicly, but I don't enjoy, nor see any point in embarrassing skeptics.
As for him taking the thread down, I have no idea why. He did it 7 hours after I told him there were no embargo issues, though that is not what he told his readers. Perhaps he hoped I would not publicly respond to his post if he had disappeared it? But I had saved a copy, and posted it on my site.
I believe he owes me an apology.
Jo
PS: We had no correspondence after mid April with him before he posted his public reply. If he'd read Davids response of April 11, he would have realized most of his post was attacking a strawman. In his last email to us he asked if David realized all solar cycles were not exactly 11 years (did he really think we were so stupid?) and asked if David believed in Intelligent Design.
Jo
There are many sceptics that have invested time in the CO2-centric theory instead of looking at it afresh. Sceptic blogs are full of admonishments to those who just don't buy it to stop hurting the sceptic cause. Accept the physics and argue the sensitivity is the advice offered and sometimes demanded. The truce position of 'we agree that we don't know' is the result. There are both sceptics and alarmists that will have to shift their positions should cooling occur and it is likely that some will find that process or even the thought of it, uncomfortable. I've just read Vlll, the conclusions of which are no surprise to solar-centric theory but the prediction that it contains and which is diametrically opposed to the current one should lead to falsification of one or the other in my lifetime. I thank you and David for that.
Ian