Please note: You are viewing the unstyled version of this web site. Either your browser does not support CSS (cascading style sheets) or it has been disabled.

Environmental Science

 

 



"Skating on Thin Ice"
Click here for more information on this research project.

Investigations into the predictability of global and limited area ocean models, using the BLUElink ocean forecasting system

Commander Robert Woodham
Director of Meteorology and oceanography from the
Royal Australian Navy and ADFA

Abstract


Small errors in the initial conditions of an oceanic model can grow rapidly, and destroy the model's forecasting skill. A deterministic forecast must therefore be considered incomplete unless accompanied by an estimate of the flow-dependent predictability. In order to introduce this concept of predictability, the seminar will begin with a review of stability theory, and its application to error growth in, and predictability of, global and limited area numerical models of the ocean. Expected error growth characteristics will be presented, and methods for identifying the fastest-growing error patterns explained.

Results will then be presented from an application of one such method, using Empirical Orthogonal Functions (EOFs) to approximate the Jacobian matrix, and hence identify the fastest growing singular vectors, to a limited area model set up over an oceanic domain in the vicinity of Perth. The dependence of the method on the norm used to measure error growth will be presented. The limited area model used for these studies is the 'Sparse Hydrodynamic Ocean Code' (SHOC), which is in routine use by the Royal Australian Navy to provide high-resolution ocean forecasts over areas of interest to the Australian Defence Force. SHOC is forced at its open boundaries by the 'BLUElink' global ocean model. The seminar will conclude with some intended avenues for future study.

[Back to top]

 

Copyright & Site information

  • CRICOS Provider No 00002J, ABN 90 952 801 237
  • Last Updated: December 1, 2008
  • Authorised by: Dr. Paul Hesse