Iowahawk takes a break from satire for a serious post explaining how to build your own hockey stick at home. Really.
2 thoughts on “Fables Of The Reconstruction”
Comments are closed.
Iowahawk takes a break from satire for a serious post explaining how to build your own hockey stick at home. Really.
Comments are closed.
Having not paid all that much attention to the analysis methods that were used in Mann’s research before, this was highly enlightening. The thing that strikes me immediately is the key assumption that a linear multiple regression model linking the observed data and the proxy variables, over a limited time scale, is appropriate. In my experience with statistical data analysis, this is the kind of analysis you do when you’re first starting out and have no idea at all of the physical model underlying the phenomena you’re trying to explain; these kind of models have very little predictive power. Of course, the complexity of climate makes it much, much more difficult to calibrate a model, but that’s life, and I don’t see how the science can possibly be settled at this point.
When I was in college, one of my professors used to count points off of tests and homework problems if we used more significant figures in our answers than called for by the data. Later on, this kind of thing fell under the pithy comment “drinking your own bathwater.” I think the AGW people are drinking plenty of their own bathwater…
The part that still amazes me is entirely in the error handling.
The base error of a generic mercury thermometer is around 0.1C.
But that’s the error for measuring temperatures immediately adjacent to the thermometer! (Ignoring the wet/dry bulb issues.)
The available thermometers are considered “the average gridcell temperature.” Nevermind the raw size of a gridcell (or the microsite issues of UHI effects.) But the average error of a stock thermometer being used to measure something fifty miles away isn’t 0.1C. It just isn’t.