NARMAX Based MLT Dependent Forecast of Energetic Electrons at GEO
Abstract
Nonlinear AutoRegressive Moving Average models with eXogenous inputs (NARMAX) represents one of the most a robust methodologies for identification of complex dynamical systems. Sheffield SNB3GEO forecasting tool has been developed a few years ago and provides online (http://ssg.group.shef.ac.uk/ssg2013/UOSSW/2MeV_EF.html) forecast of the daily averaged fluxes of high energy electrons at GEO. Here we presenting a new forecasting model for energetic electron with increased temporal and spatial resolution in comparison to the SNB3GEO forecast of daily flux of energetic electrons averaged over the GEO orbit. This newly developed model provides hourly forecast for each MLT segment of GEO. As the model is newly developed only "past cast" performance could be used to estimate its prediction efficiency. It is shown that the prediction efficiency of the new model is very similar to the SNB3GEO forecast.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMSM31D3529B
- Keywords:
-
- 1942 Machine learning;
- INFORMATICSDE: 7924 Forecasting;
- SPACE WEATHERDE: 7959 Models;
- SPACE WEATHERDE: 7999 General or miscellaneous;
- SPACE WEATHER