New Insights on Geomagnetic Storms from Model Simulations Using Multi-Spacecraft Data

The forecast of the terrestrial ring current as a major contributor to the stormtime Dst index and a predictor of geomagnetic storms is of central interest to ‘space weather’ programs. We thus discuss the dynamical coupling of the solar wind to the Earth's magnetosphere during several geomagnetic storms using our ring current-atmosphere interactions model and coordinated space-borne data sets. Our model calculates the temporal and spatial evolution of H⁺, O⁺, and He⁺ ion distribution functions considering time-dependent inflow from the magnetotail, adiabatic drifts, and outflow from the dayside magnetopause. Losses due to charge exchange, Coulomb collisions, and scattering by EMIC waves are included as well. As initial and boundary conditions we use complementary data sets from spacecraft located at key regions in the inner magnetosphere, Polar and the geosynchronous LANL satellites. We present recent model simulations of the stormtime ring current energization due to the enhanced large-scale convection electric field, which show the transition from an asymmetric to a symmetric ring current during the storm and challenge the standard theories of (a) substorm-driven, and (b) symmetric ring current. Near minimum Dst there is a factor of ∼ 10 variation in the intensity of the dominant ring current ion specie with magnetic local time, its energy density reaching maximum in the premidnight to postmidnight region. We find that the O⁺ content of the ring current increases after interplanetary shocks and reaches largest values near Dst minimum; ∼ 60% of the total ring current energy was carried by O⁺ during the main phase of the 15 July 2000 storm. The effects of magnetospheric convection and losses due to collisions and wave-particle interactions on the global ring current energy balance are calculated during different storm phases and intercompared. This revised version was published online in August 2006 with corrections to the Cover Date.