Physics
Scientific paper
May 2002
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2002agusmsm42b..06u&link_type=abstract
American Geophysical Union, Spring Meeting 2002, abstract #SM42B-06
Physics
2788 Storms And Substorms, 3220 Nonlinear Dynamics, 3240 Chaos
Scientific paper
The solar wind - magnetosphere coupling during substorms exhibit both global end multi-scale features. Local-linear filters (LLF) with autoregression are used to model the magnetospheric dynamics during substorms. The filters are derived from the reconstructed input-output magnetospheric phase space using time series of VBs as the input and AL index as the output of the model. LLF reveal both globally coherent and multi-scale features of dynamics. It is shown that for the purpose of long-term forecasting the conventional formula for LLF can be greatly simplified. The filter response can be estimated in a mean-field fashion, i.e. averaging outputs corresponding to the similar states of the system in the reconstructed phase space. The mean-field method allows accurate iterative predictions of AL time series for several days in a row during periods of both high and low magnetospheric activity, capturing large-scale variations the best. Being regular and predictable, the part of AL reproduced by the mean-field model corresponds to the coherent component of substorm-time magnetospheric dynamics. The part of AL time series not captured by the simple mean-field model is multi-scale in nature, i.e. it consists of perturbations of all scales and has a power spectrum similar to colored noise. It is demonstrated that this remainder can be reproduced by the simplified version of LLF if the dimensionality of the embedding space is varied at each step of iterative predictions. This procedure yields a factor of four improvement in the accuracy compared to the mean-field model. The purpose of current study is to obtain the local dimensionality of the embedding space using the input data which would then allow accurate forecasting of both global and multi-scale aspects of magnetospheric dynamics.
Mikhail M.
Papadopoulos Demetrios
Sharma Sanjeev
Ukhorskiy Aleksandr Y.
No associations
LandOfFree
Modeling and Forecasting the Multi-Scale Constituent of Magnetospheric Dynamics during Substorms with Local-Linear Filters does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Modeling and Forecasting the Multi-Scale Constituent of Magnetospheric Dynamics during Substorms with Local-Linear Filters, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Modeling and Forecasting the Multi-Scale Constituent of Magnetospheric Dynamics during Substorms with Local-Linear Filters will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1725219