Home » Subbotin and Shprits 2009

Three-dimensional modeling of the radiation belts using the Versatile Electron Radiation Belt (VERB) code

Subbotin D. A., Y. Y. Shprits, (2009), Three-dimensional modeling of the radiation belts using the Versatile Electron Radiation Belt (VERB) code, Space Weather, 7, doi:10.1029/2008SW000452

Abstract

Relativistic electrons are hazardous for satellites and humans in space. In this study, we present a detailed description of the Versatile Electron Radiation Belt (VERB) code. The computationally efficient methods described in this report make the VERB code a useful tool for space weather forecasting and nowcasting of the relativistic electron environment. The computational efficiency of the code makes it also appropriate for future use with data assimilation tools to specify the state of the radiation environment and correct imperfect models. We also present several skill scores which can be used to quantify the state of the relativistic electron environment. Our numerical approach is based on the use of two grids: one for radial diffusion and the other for energy and pitch angle diffusion. In this paper, we describe the initial and boundary conditions, the grids and time step used for our simulations, and the tests we conduct to verify the validity of our numerical approaches. Specifically, we perform simulations with time steps ranging from 0.01 to 4 h and number of grid points ranging from 7 to 76 for each of the variables. We compare the results obtained using various spatial resolutions and time steps, giving the relative solution errors for each of the performed simulations on the basis of the developed set of skill scores. The choice of a 0.1 h time step with a grid resolution of 19 × 19 × 19 points was found to be optimal. We compare the block and split methods, showing that the split method is much faster than the block method and sufficiently accurate for time steps of half an hour or less. Additionally, we present an approximate method that simplifies simulation by using only one grid and compare it to the more accurate two-grid method.

Authors (sorted by name)

Shprits Subbotin

Journal / Conference

Space Weather

Acknowledgments

This research was supported by NSF GEM grant ATM‐0603191 and AFRL grant FA9550‐08‐1‐0140.

Grants

ATM‐0603191 FA9550‐08‐1‐0140

Bibtex

@article{doi:10.1029/2008SW000452,
author = {Subbotin, D. A. and Shprits, Y. Y.},
title = {Three-dimensional modeling of the radiation belts using the Versatile Electron Radiation Belt (VERB) code},
journal = {Space Weather},
year = {2009},
volume = {7},
number = {10},
pages = {},
keywords = {modeling, radiation belts, VERB code},
doi = {10.1029/2008SW000452},
url = {https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2008SW000452},
eprint = {https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2008SW000452},
abstract = {Relativistic electrons are hazardous for satellites and humans in space. In this study, we present a detailed description of the Versatile Electron Radiation Belt (VERB) code. The computationally efficient methods described in this report make the VERB code a useful tool for space weather forecasting and nowcasting of the relativistic electron environment. The computational efficiency of the code makes it also appropriate for future use with data assimilation tools to specify the state of the radiation environment and correct imperfect models. We also present several skill scores which can be used to quantify the state of the relativistic electron environment. Our numerical approach is based on the use of two grids: one for radial diffusion and the other for energy and pitch angle diffusion. In this paper, we describe the initial and boundary conditions, the grids and time step used for our simulations, and the tests we conduct to verify the validity of our numerical approaches. Specifically, we perform simulations with time steps ranging from 0.01 to 4 h and number of grid points ranging from 7 to 76 for each of the variables. We compare the results obtained using various spatial resolutions and time steps, giving the relative solution errors for each of the performed simulations on the basis of the developed set of skill scores. The choice of a 0.1 h time step with a grid resolution of 19 × 19 × 19 points was found to be optimal. We compare the block and split methods, showing that the split method is much faster than the block method and sufficiently accurate for time steps of half an hour or less. Additionally, we present an approximate method that simplifies simulation by using only one grid and compare it to the more accurate two-grid method.}
}