Press release of the week

>>

by Naomi Bloch

Kudos to McGill's University Relations Office for shrouding a potentially earth-shattering news bulletin in a haze of gobbledygook.

It would seem that McGill physicist Shaun Lovejoy and his colleagues have debunked the supremacy of modern weather forecasting techniques. How, you might ask? According to the press release: "His work, to be published in Physical Review Letters, shows that the dynamics of the atmosphere from planetary scales down to 1 km, at least, are extremely close to those predicted by scale invariant but stratified cascade processes. 'This demonstrates--in the most intensive data study to date--that the atmosphere is not divided into isotropic three-dimensional and isotropic two-dimensional turbulent ranges,' says Lovejoy. 'Instead, the large structures are on average the same as the smaller ones except for a progressive flattening at larger scales; they are special types of (multi) fractals.'

"Numerical weather (i.e. computer-based weather forecasting) and climate models are justified with a qualitatively different model of the atmosphere, in which the small structures (gusts, "turbulence" etc.) are quite different--and separated from--the large scale structures traditionally associated with 'weather.' Because of this scale separation, says Lovejoy, turbulence can be considered as no more than small unimportant perturbations on the weather."

If that doesn't clear it up for you, you'll just have to call up Dr. Lovejoy yourself.