| Volumes and Issues Contents of Issue 3 |
www.atmos-chem-phys-discuss.net/4/3653/2004/
doi:10.5194/acpd-4-3653-2004
© Author(s) 2004. This work is licensed
under a Creative Commons License.
Using an extended Kalman filter learning algorithm for feed-forward neural networks to describe tracer correlations
1Global Modelling and Assimilation Office, NASA Goddard Space Flight Center, USA
2GEST at the University of Maryland Baltimore County, MD, USA
3Unilever Cambridge Centre, Department of Chemistry, University of Cambridge, United Kingdom
Abstract. In this study a new extended Kalman filter (EKF) learning algorithm for feed-forward neural networks (FFN) is used. With the EKF approach, the training of the FFN can be seen as state estimation for a non-linear stationary process. The EKF method gives excellent convergence performances provided that there is enough computer core memory and that the machine precision is high. Neural networks are ideally suited to describe the spatial and temporal dependence of tracer-tracer correlations. The neural network performs well even in regions where the correlations are less compact and normally a family of correlation curves would be required. For example, the CH4-N2O correlation can be well described using a neural network trained with the latitude, pressure, time of year, and CH4 volume mixing ratio (v.m.r.). The neural network was able to reproduce the CH4-N2O correlation with a correlation coefficient between simulated and training values of 0.9997. The neural network Fortran code used is available for download.
Discussion Paper (PDF, 698 KB) Interactive Discussion (Closed, 3 Comments) Publication in ACP not foreseen
Citation: Lary, D. J. and Mussa, H. Y.: Using an extended Kalman filter learning algorithm for feed-forward neural networks to describe tracer correlations, Atmos. Chem. Phys. Discuss., 4, 3653-3667, doi:10.5194/acpd-4-3653-2004, 2004. Bibtex EndNote Reference Manager XML
