To The Sun And The Moon

Over at the NZ CSC, I've been chasing Ken to provide me with details of three studies that he says support his claimed forecast accuracy. A while ago (Oct 13) he posted this:

"A prominent secondary school team did a science project that studied me for 3 months at a MOTAT exhibition sponsored by NIWA. They came up with 87%. A Massey University project gave me 85-91%. Auspacwx ran a Random Number Generator against me, and gave me above 80%. I am happy with 85% and it seems to be average. Metservice give themselves about the same most of the time."

I did some Googling and found this site, where Aussie weather follower Carl Smith took Ken's forecasts for Australia (provided by Ken to the austpacwx discussion list) and compared them to what actually happened. Carl's exhaustive effort showed that Ken was doing quite well for the first half of the month, but then the wheels fell off, and under Carl's scoring system he ended up with about 50% - more or less the same as my rainfall and sunshine analysis.

I've since been chasing Ken to provide details of the "Massey University study" in which he did so well, but he seems reluctant to provide any data. The school science fair project is similarly mysterious. Following up on Carl Smith's work, I found another exercise Carl had conducted, this time examining how well several noted astrometeorologists did on predicting the weather over the Sydney Olympics. Ken's work was included, and he was second best in show - scoring 60% on Carl's system, behind US forecaster Carolyn Egan, whose work we noted earlier [link].

To be fair to Carl (and to Ken), I should note that Carl believes there may be something to the forecast methods used by Ken, but that it needs refinement. In an email to me, copied to Ken and which Ken decided to post at the NZ CSC he said; "I do think Ken is on to something with the Moon as a factor in long range forecasting, however I also think Ken has more work to do with regards to sorting out just what contribution the Moon makes, and how and where that plays out in the real world." Carl also provided me with links to two papers (here, here)[links] by Charles Keeling & Timothy Whorf of the Scripps Institute in California, who postulate a relationship between long term lunar and solar tidal cycles and climate. Unfortunately for Ken, they explicitly rule out the mechanism he proposes drive the weather (atmospheric tides): "We focus on oceanic tides, because the atmospheric tides are too small to be of practical importance in comparison" (Keeling & Wharf, 1997).

Meanwhile, I'm still waiting for details of the Massey "study". I'll pass on the school science fair project. Having helped out with one or two of those in my time, I know how interesting they can be. But I wouldn't stake my career on the accuracy of their results.