Physics
Scientific paper
Dec 2009
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2009agufmsm54a..01m&link_type=abstract
American Geophysical Union, Fall Meeting 2009, abstract #SM54A-01
Physics
[2722] Magnetospheric Physics / Forecasting, [2753] Magnetospheric Physics / Numerical Modeling, [7959] Space Weather / Models, [7959] Space Weather / Models
Scientific paper
A variational data assimilation algorithm has been developed to incorporate electron flux time-series data from satellites into a simple one dimensional diffusion model for the radial transport of radiation belt electrons. The model developed assumes a power law scaling for the radial diffusion coefficient with L shell. The effectiveness of this method is investigated by means of a series of identical twin numerical experiments. This involves using the diffusion model to produce synthetic observations along various satellite trajectories. These observations are in turn used to estimate time-dependent parameters input to the diffusion model, which are compared against the values initially used. The data assimilation algorithm considers the time dependent source located at the outer boundary as a function to be determined. Using synthetic satellite electron flux observations, the algorithm computes a source function that, when used as an input to the diffusion model, most closely reproduces the synthetic observations in a least-squares sense. Observational errors are taken into account, and an estimate of the uncertainty in the output source function is also produced. This uncertainty is found to consistently reflect the quality of the source function estimation during identical twin numerical experiments. Initial tests indicate that the quality of the outer boundary source estimation is strongly dependent on the satellite location, indicating that the outer boundary source estimation becomes poor as information relating to the outer boundary contained in the observations is reduced. The potential of using this data assimilation method to estimate one or more parameters that determine the radial diffusion coefficient, and the possibility of determining whether physical processes affecting the observations are missing in the dynamical model will be discussed.
Degeling A. W.
Kabin Konstantin
Marchand Régine
O'Donnell Shawn
Rankin Robert
No associations
LandOfFree
Data Assimilation Using a Variational Method for a 1D Radiation Belt Diffusion Model 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 Data Assimilation Using a Variational Method for a 1D Radiation Belt Diffusion Model, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Data Assimilation Using a Variational Method for a 1D Radiation Belt Diffusion Model will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1881832