Comparison of three methods of forecasting 24-hour tropical cyclone movement in the Western Australian region

Hopwood, J. and Scott, W.D. (1992) Comparison of three methods of forecasting 24-hour tropical cyclone movement in the Western Australian region. Journal of the Royal Society of Western Australia, 75 . pp. 3-7.

Abstract

Three methods of forecasting tropical cyclone tracks in the Western Australia region are discussed. One is a version of Neumann's CUPER adapted for the region and referred to as CPLR, another is a method developed in Western Australia which uses the low-level relative vorticity field in the environment of the cyclone as a predictor, and the last is pure persistence. The three methods were used to produce 214 triplets of forecasts from best-track data for 13 cyclones and the results compared. The Wilcoxon test was used to test the significance of differences between forecasts for various subsets of cyclones. The results suggest that (a) the performances of the first two methods match that of persistence overall, and are superior when the error in the persistence forecast is greater than the mean error or when the cyclones are recurving, and (b) the second method matches CPLR overall and surpasses it with some categories of cyclones.

Item Type:	Journal Article
Murdoch Affiliation(s):	School of Biological and Environmental Sciences
Publisher:	Royal Society of Western Australia
Copyright:	© Royal Society of Western Australia
URI:	http://researchrepository.murdoch.edu.au/id/eprint/17640

Item Control Page

Downloads