Premium
On the modelling of over‐ocean hurricane surface winds and their uncertainty
Author(s) -
Khare S. P.,
Bonazzi A.,
West N.,
Bellone E.,
Jewson S.
Publication year - 2009
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1002/qj.442
Subject(s) - predictability , meteorology , planetary boundary layer , empirical orthogonal functions , covariance , boundary layer , parameterized complexity , residual , wind speed , range (aeronautics) , environmental science , field (mathematics) , log wind profile , wind direction , climatology , geology , mathematics , statistics , geography , turbulence , algorithm , wind gradient , physics , aerospace engineering , engineering , mechanics , pure mathematics
This paper investigates the modelling of over‐ocean hurricane surface wind fields and their associated uncertainty. The wind models tested include parameterized balance models, a two‐dimensional numerical planetary boundary‐layer model and a three‐dimensional (3D) linear analytical boundary‐layer model. Using a set of archived over‐ocean surface wind field reconstructions for validation, a series of cross‐validation experiments has been performed for a range of norms. For norms that quantify predictability of vector fields, a particular configuration of the 3D analytical model was found to be superior to the other models tested. Using residual fields derived from fitting the wind models to the validation data, the issue of how to model the uncertainty (in the form of a covariance) in the speed field has also been examined. Covariance models based on truncated empirical orthogonal function representations were found to be optimal. Copyright © 2009 Royal Meteorological Society