Date of Award

Fall 2007

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

Vania K Jordanova


An improved version of the ring current-atmosphere interactions kinetic model (RAM) is presented in this thesis. The recent stormtime empirical model T04s and the IGRF model are used to represent the Earth's external and internal magnetic fields respectively. Particle drifts, losses due to charge exchange with geocoronal hydrogen and atmospheric losses are included in the model as they are considered the main mechanisms of ring current development and its following decay. A numerical technique for bounce-averaging along the field lines is introduced and results for the calculated bounce-averaged hydrogen densities and magnetic gradient-curvature drift velocities (general case) for the moderate storm of April 21-25, 2001, are presented. A comparison in the calculations between T04s and a dipole field shows that the bounce-averaged hydrogen density for T04s differs with ∼ 5% from that for a dipole field for quiet time and it may become 30% smaller for disturbed conditions on the nightside for L > 4. The gradient-curvature velocities for T04s at large L-shells are ∼ 20% higher on the nightside and 20% lower on the dayside than those for a dipole field for quiet time. For disturbed conditions they are respectively ∼ 200% higher and 20% lower than the dipole values. The contribution of the cross-B term to the magnetic drift is ∼ 5%. Results for the time evolution of the trapped equatorial flux for H+, He+, and O+ ions for various particle energies and pitch angles obtained by the new model with a non-dipole field (RAM-ND) are presented. The new computations for the April 2001 storm using a Volland-Stern convection model show a slight continued increase in the flux and the total ring current energy for the three ion species even after the storm main phase. A higher increase in the flux is observed towards the dusk side for the RAM-ND model compared to RAM due the difference in the charge exchange rates and the azimuthal drifts for the two different geomagnetic field configurations. Both models give similar values for the low energy ion fluxes. The high energy component of the ion flux for large pitch angles for RAM-ND has more strongly expressed dominance during the storm recovery phase. The increase in the O+ flux after the storm minimum Dst given by RAM-ND, indicates a continuous several hour activity of the various ionospheric sources during stormtime, leading to the accumulation of energetic O+. The contribution of He + to the total ring current is about 4%. The total ring current energy using RAM-ND is reduced by ∼ 30% compared to RAM. The results obtained by the RAM-ND model confirm recent calculations by other models and they are consistent with previous satellite measurements. The energy spectra of the calculated spin averaged ion flux is compared with Polar/CAMMICE-MICS data few hours before the Dst minimum. The flux profile dependence on the L-shell value is studied at the midnight-dusk and at the prenoon side. Both RAM-ND and RAM fluxes predict the measurements beyond 3 R E reasonably well with the RAM-ND model performing slightly better than RAM in the mid-energy range. At lower L-shells the measured low energy flux does not have a well expressed minimum, while the modeled fluxes for both models have deep and broad minima. The dip in the modeled flux profiles at all L-shells is shifted towards higher energies compared to the dip in the data. The minima shift for RAM-ND is smaller than the minima shift for RAM in most of the cases. An approximate calculation of the perturbation in the Dst due to the change in the total stormtime ring current energy content is presented. The depression in the Dst for RAM-ND is about two times smaller than the change predicted by RAM due to the smaller total energy for RAM-ND.