|2012-02-20 11:20:45||New paper predicting imminent cooling due to solar cycles|
Got this email:
Seems to be a new paper in arxiv which is the denier publishing house du jour lately
In the conclusions:
"Our forecast indicates an annual average temperature drop of 0.9◦C in
The stats is beyond me - I'd be tempted to see if Tamino would take a look at this.
Also note they have done a bit with David Archibald.....
Quite a few typos in there too....
Cheers - John
1) From the abstract:
Those figures are certainly consistent with greater than 50% of warming post 1950 being the consequenc of anthropogenic causes. So, much as they may desire it, and as Humlum is an author, they do desire it, their evidence does not contradict the IPCC.
2) There figure 1 showing solar cycle length is a poor match to the global temperture series, even if lagged by eleven years. As I believe a study has already been done showing the supposed correlation between solar cycle length and GMST does not hold post 1970 (can't remember name and author unfortunately), their failure to discuss that lack of correlation, and to point out that this is a regional effect if that, is a glaring omission.
3) As I am sure Tamino points out, by going regional they vastly improve the chances of a coincidental match by vastly increasing their sample size. Inferences from this result, even if valid, to global causes is therefore very much unwarranted; and
4) The charts such as their figure 1 always puzzle me. Purportedly each point represents a solar cycle, and the y-axis indicates the length of the cycle. It follows that the seperation along the x axis of any two points is given by the either one of the two values on the y axis (if lagged) or the average of the value on the y axis (if the point is at the midpoint of the cycle). Yet the y axis values of the first six points are in order:
10; 8.5; 14; 11.5; 10.5; 11
And the seperation of the first five points on the x axis are in order:
8.3; 11; 11.6; 13.6; 9.8
Those respective values seem to me to be strictly inconsistent. I am not an astronomer, so astronomers may have some wierd naming convention that resolves this inconsistency. Never-the-less, the point is bizzare.
I saw this during last week in a real journal. It was Journal of Solar and Terrestrial physics, I believe, a journal that seems to publish a constant stream of these denier flavour articles. Also, one of the authors is Ole Humlum, who has been publishing this kind of stuff (and even worse) before.
Oh the Humlum paper. I think it's essentially the same as the Archibald paper, in fact I believe the paper has high praise for Archibald. The whole solar cycle length argument has been debunked for over a decade, and of course they're relying on local temperatures (like Archibald does). Both are just a joke.