Please enable JavaScript to view the comments powered by Disqus. blog comments powered by Disqus

A Lesson in Graphing and Trend Fitting from WTFUWT

Steven Goddard, as loyal readers of this blog will already know, has transformed Venusian atmospheric physics with his egregious (that word as used here means standing out from the flock not the derogatory unusually crappy) thinking on adiabatic heating, thereby rubbishing the greenhouse effect theory in our solar system.

Mr Goddard has now turned his penetrating insight onto that favourite bête blanche of the alarmists, the supposedly shrinking Arctic sea ice. Goddard performs some simple time series analysis on the JAXA data, disentangling all that confusing varicoloured spaghetti and shows how the long term trend is revealed by a linear fit. Thus:

wuwt godd

(Gentle readers of the FoGT blog may be confused by the last sentence in Goddard’s caption where he uses the rhetorical device of irony, whereby the literal meaning of his words is the opposite of what he really means to say.) Some critics immediately jumped on Goddard’s results and claimed that the California-sized increase in sea ice was an artifact of Goddard’s analysis and that everything depends on where you pick the starting point. I suppose that, taking this idea to an extreme, means that the JAXA data could even show “more proof that the Arctic is melting down”, which would be doubly ironic when you think about it.

FoGT scientists have done their own analysis using a model data set known to us mathematicians as the “sine” function.

jamarr sine

This graph is remarkably similar to Goddard’s graph (if you ignore the “noise” with the unusually low minima in recent years) and shows the same increasing linear trend discovered by him. The lesson here is that when you analyze cyclic data like climate, you have to take care to do your analysis over many cycles and to pick your start and end points carefully. For example, using a time series shifted by half a cycle (that’s six months for you astronomical buffs) you get the following graph, which would produce the patently absurd result that Arctic ice is actually shrinking. Tell that to the Navy!


As Steven Goddard so succinctly--if ungrammatically--put it: Linear trends through cyclical data forms the entire basis of the IPCC’s raison d’être.

Apart from those non-linear hockey sticks, of course.