 |
| Titel |
How system identification techniques can help in the study of relativistic electrons and ions at GEO |
| VerfasserIn |
Richard J. Boynton, Michael A. Balikhin, Stephen A. Billings |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250079996
|
|
|
|
|
|
| Zusammenfassung |
| The standard scientific approach is to develop mathematical models from first physical
principles. However, there are many complex dynamical systems where a mathematical
model cannot be deduced from first principles with our present level of knowledge, such as
the radiation belts. System science techniques, such as the NARMAX algorithm, are able to
automatically determine the dynamical equations that govern the evolution of the complex
system from input-output data. From the physically interpretable models of the NARMAX
algorithm, it is possible, in some sense, to reverse engineer and obtain understanding of the
physical object and the processes involved. Here, the NARMAX system science approach is
applied to the evolution of the radiation belts, where ACE measurements are used as
the inputs and daily averaged GOES and LANL particle flux data are considered
as the outputs. The Error reduction ratio (ERR), a key concept of the NARMAX
algorithm, is employed to assess the solar wind control parameters of the particle
fluxes. For low energies, the electron fluxes are shown to be influenced by the solar
wind velocity with a dependance between the time delay of the velocity and the
energy of the electrons. For high energies, it shown that the solar wind density
becomes the most significant control parameter. The ion fluxes are mainly effected by
the solar wind velocity but also the energetic ions in the solar wind. An online
real time forecasting tool has been developed from the results of the NARMAX
approach that provide reliable estimates of the >800 keV and >2 MeV electron
fluxes. |
|
|
|
|
|
|