Answers are grounded in this paper and cite their sources.
Statistical models predict wind behavior in Canterbury, suggesting strong gusts during cascade events.
We derive practical forecasting formulae from statistical models relating northwesterly surface wind on the Canterbury Plains, New Zealand, to concurrent conditions over Canterbury and the Southern Alps to the west. The most important term in models of maximum gust was a time-variation of the magnitude of pressure difference across the Alps, motivated by high-resolution wind spectra from the Canterbury Plains. We find that the isallobaric wind does not explain this effect, and furthermore that an isallobaric wind term is insignificant and can be omitted. Spectra of northwesterly wind at seven locations in a line on the Canterbury Plains were computed from wind samples at 20-second intervals. Many of these spectra follow a -5/3-power line to periods of more than an hour. We suggest that this is evidence for large-scale three-dimensional (though not isotropic) eddy cascades with fluctuations that cannot be predicted from large-scale concurrent conditions over the Southern Alps. During a cascade event, the variation of wind would then be unpredictable on time-scales of one hour and less, which is a substantial component of the total variation. We further suggest that the strongest surface wind is likely to occur during such cascade events. Global model analyses at standard pressure levels from the European Centre for Medium-Range Weather Forecasts (ECMWF), and the observed pressure difference across the Alps together with a measure of its time-variation, were used as predictors of observed daily maximum three-second gust speed and 10-minute mean wind speed. The models of maximum gust speeds were found to be more accurate than the models of 10-minute mean wind speed, but both accounted for roughly 50 per cent of the variance of the speed. The models also accounted for about 50 per cent of the variance of the square of the speed (which is proportional to aerodynamic load), so that they are suitable for predicting severe wind. The model for dry conditions accounted for 63 per cent of the variance of daily maximum gust speed at Christchurch airport, and 60 per cent of the variance of daily maximum aerodynamic load.
No takes yet. Share an insight, caveat, or question.
Purnell et al. (1999) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: