 |
| Titel |
Progress towards the adjoint dynamo problem |
| VerfasserIn |
Kuan Li, Andrew Jackson, Philip Livermore |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250047558
|
|
|
|
|
|
| Zusammenfassung |
| Data assimilation is a powerful tool for understanding the evolution of a time-varying
physical system. For example, it can be used to determine properties of control parameters or
initial conditions. This method is widely applied in environmental science, meteorology and
seismology. A most important aspect of the analysis involves deriving and computing the
adjoint model.
In the geodynamo problem, governing the evolution of the Earth’s magnetic field, the
material properties of the core are assumed homogeneous and known, and therefore our
interest is in determining the initial 3-D magnetic field configuration. In this system,
constraints are provided by a wide spectrum of observations, from very accurate satellite data
recorded over the last few decades, to archeomagnetic observations over the last ten thousand
years.
As an initial study, we created a mathematical framework and the corresponding
numerical algorithm for computing the adjoint of the magnetic induction equation and
applied it to the inversions of a kinematic dynamo model and a 3-D nonlinear toy
model. We analyse the dependency of the results on factors such as the magnetic
Reynolds number and the length of the assimilated observational time window. |
|
|
|
|
|
|