The relativistic electron response at geosynchronous orbit during the January 1997 magnetic storm

The first geomagnetic storm of 1997 began on January 10. It is of particular interest because it was exceptionally well observed by the full complement of International Solar Terrestrial Physics (ISTP) satellites and because of its possible association with the catastrophic failure of the Telstar 401 telecommunications satellite. Here we report on the energetic electron environment observed by five geosynchronous satellites. In part one of this paper we examine the magnetospheric response to the magnetic cloud. The interval of southward IMF drove strong substorm activity while the interval of northward IMF and high solar wind density strongly compressed the magnetosphere. At energies above a few hundred keV, two distinct electron enhancements were observed at geosynchronous orbit. The first enhancement began and ended suddenly, lasted for approximately 1 day, and is associated with the strong compression of the magnetosphere. The second enhancement showed a more characteristic time delay, peaking on January 15. Both enhancements may be due to transport of electrons from the same initial acceleration event at a location inside geosynchronous orbit but the first enhancement was due to a temporary, quasi-adiabatic transport associated with the compression of the magnetosphere while the second enhancement was due to slower diffusive processes. In the second part of the paper we compare the relativistic electron fluxes measured simultaneously at different local times. We find that the >2-MeV electron fluxes increased first at noon followed by dusk and the? dawn and that there can be difference of two orders of magnitude in the fluxes observed at different local times. Finally, we discuss the development of data-driven models of the relativistic electron belts for space weather applications. By interpolating fluxes between satellites we produced a model that gives the >2-MeV electron fluxes at all local times as a function of universal time. In a first application of this model we show that, at least in this case, magnetopause shadowing does not contribute noticeably to relativistic electron dropouts.