SM33B-01 13:40h
Model and Data Comparison of Inner Magnetosphere Plasma
The previous study of comparison between Lyon-Fedder-Mobbary (LFM) MHD simulations and GOES data shows that the MHD simulations overestimate the magnetic fields during storm main phase especially at the nightside (Huang et al., 2003, Fall AGU meeting). The same result is shown by comparing the simulation with Polar Magnetic Field Instrument at different storm phases. To understand how the plasma distributions modify the field line configuration, we compare the particle pressure data from Polar Hydra and CAMMICE instruments with the MHD simulations. The Polar data contain full coverage of the inner magnetosphere integrated over time from 1996 to present. We use these data to construct 3-D binned plasma maps of the inner magnetosphere as a function of state parameters, such as solar wind conditions, Dst index and their time histories. This study is complementary to work done by Ebihara (2002) who used Polar MICS data to explore the storm time proton ring current. Establishing the distributions of plasma and current systems as functions of magnetospheric conditions, the goal of this study, will provide a general resource for other data/model comparisons and will ultimately help in developing higher fidelity inner magnetosphere models.
SM33B-02 13:55h
Simulations of Stormtime Ring-Current Magnetic Field Produced by Ions and Electrons
We simulate the three dimensional structure of the ring-current magnetic field by tracing the guiding centers of representative ions and electrons from the plasma sheet as they bounce between mirror points and drift across {\bf B} in a model magnetosphere. The ambient magnetic field model we use for this study is the Dungey model, which consists of a dipole field plus a uniform southward -tail" field. We map a spatially analytical expansion of the AMIE ionospheric electric potential, expressed as a function of magnetic latitude and magnetic local time, along magnetic field lines (for {\it L} $\ge 2$) throughout this model magnetosphere. We trace bounce-averaged drifts for ions and electrons conserving their first two adiabatic invariants m and J (with values that correspond to energies $\sim 10 - 300$ keV at {\it L} = 3). Using these simulation results, we map phase space densities according to Liouville's theorem but taking into account losses due to charge exchange for protons and losses due to wave-particle interactions for electrons. We specify an initial proton ring current distribution by solving the steady-state transport equation that balances quiescent radial diffusion against charge exchange. To obtain MLT-dependent and UT-dependent boundary values for our phase space density distribution, we map geosynchronous LANL particle data to the boundary of our model magnetosphere. From the simulated phase-space densities, we calculate the particle pressure and energy density distributions. From the pressure distributions, we compute the ring current magnetic field. From simulations of the 19 October 1998 storm, we find that the large AMIE electric field in the evening sector would have led to rapid ($\sim$ 20 minutes) inward transport of plasmasheet ions from the neutral line to $L \sim 3$ near the dusk meridian. We can thus account for the observed rapid formation of the partial proton ring current there and its subsequent symmetrization to a wider range of MLT. In regions where the ring current is especially intense, the ring-current magnetic field can be a significant fraction of the Earth's ambient magnetic field. This suggests a future need for eventually calculating particle transport in a magnetically self-consistent model.
SM33B-03 14:10h
Contribution of charge exchange at high altitudes to ring current decay: IMAGE/HENA observation
Several mechanisms have been proposed to explain the decay of the storm-time ring current. Although charge exchange is believed to be one of the most probable causes of the decay, there have been few observations that have been able to quantitatively estimate its contribution. In this paper, we evaluate the contribution of charge exchange process to the ring current decay, using energetic neutral atom (ENA) data ($>$ 10 keV) obtained by the HENA imager onboard the IMAGE satellite. We focus on charge exchange collisions at high altitudes (i.e., at larger radial distance within ring current L-shells) in particular. The HENA imager detects ENAs which are generated when ring-current energetic ions lose their charges through collisions with neutral atom and molecules of the upper atmosphere and exosphere. The energy of a detected neutral atom is considered to be equal to the energy lost by a ring current ion. The rate of energy loss through charge exchange within each line-of-sight of pixels can be expressed as: $dE/dt = 4\pi \int E dE \int\!\!\!\int s^2 ds d\Omega \cdot \sigma_{10} n_H j_{ION} $, where $\sigma_{10}$ is the charge exchange cross section, $n_H$ is the density of geocorona, $j_{ION}$ is the flux of ring current ions, $s$ is the distance from the satellite to the position of ENA production, $d\Omega$ is a solid angle of each line-of-sight, and $E$ is energy of ENAs, if we assume isotropic pitch angle distribution. Since $s$ needs to be determined for estimating the energy loss, we assume that all ENAs were generated at the spherical shell with a radial distance of 8 $R_E$. We calculated the energy loss rate for the recovery phase of three storms (April 22, 2001, September 23, 2001, and October 21, 2001), using both hydrogen ($>$ 10 keV) and oxygen ($>$ 50 keV) data. The IMAGE satellite was located above the North Pole (MLAT $>$ 80 degrees) and near its apogee (radial distance $>$ 7 $R_E$) during those intervals. The calculated loss rate was less than 1/10 of the decay rate of ring current ions estimated from Burton's formula (Burton et al., p.4202, JGR, 1975) for all events. The above assumption regarding the position of ENA production likely gives an overestimate, because many parts of the detected ENAs were probably produced around the magnetic equator. Therefore, our result suggests that charge exchange at radial distances greater than 3~$R_E$ hardly contributes to the decay of the ring current. We will report dependence of the energy loss rate on the recovery rate of storms, using more realistic assumptions. Energy loss at low altitudes (i.e., around footpoints of ring current L-shells) will be also discussed. Our first assumption (i.e., isotropic pitch angle distribution) is not considered valid at low altitudes at high latitudes, where ion distribution could be strongly anisotropic as a result of ions lost into the atmosphere.
SM33B-04 14:25h
Estimation of the Substorm-Related Variation of the Ring Current Intensity from the Sym-H Index
The development of magnetospheric storms has been conventionally addressed in terms of the Sym-H(Dst) index, which is ideally regarded as a measure of the ring current intensity. However, it is highly likely that a considerable fraction (~25%) of storm-time Sym-H reduction can be actually attributed to the tail current. Furthermore, the results of previous studies strongly suggest that the substorm-associated variations of the tail current and the ring current have opposite effects on Sym-H, and that the former overcompensates the latter. Thus, for the substorm timescale, using the Sym-H index as a measure of the ring current intensity is misleading with regard not only to the intensity of the ring current but also to the timing of the ring current development and decay. In the present study we address this issue by decomposing storm-time variations of Sym-H into {\it low-frequency} and {\it high-frequency} components. It is inferred that the {\it low-frequency} component is driven mostly by magnetospheric convection, whereas the {\it high-frequency} component includes most of substorm-related variations. Using the total energetic neutral atom (ENA) flux measured by the IMAGE/HENA instrument as a proxy of the ring current intensity, we quantitatively address how this {\it high-frequency} component is related to the variation of the ring-current intensity.
SM33B-05 14:40h
Modeling the Inner Radiation Belt Protons
The main sources of energetic protons for the inner zone of the radiation belt are CRAND (cosmic ray albedo neutron decay) and inward diffusion of solar protons. In the absence of detailed measurements from the inner zone, a theoretical model can be of practical value in specifying the CRAND proton intensity, which is the dominant component at the higher trapped proton energies. Calculations done in the 1960s and '70s verified that CRAND can be a significant proton source. With updated model inputs and modern computers we now have the potential to calculate accurate intensities as a function of the three adiabatic invariants of trapped particle motion and of solar cycle phase. Significant aspects of the modeling include: calculation of the albedo neutron flux from models of the galactic cosmic rays and neutral atmosphere, drift averaging of the neutron decay source and the atmospheric densities, numerical calculation of adiabatic trapping limits in the geomagnetic field, and integration of the model transport equation with time and energy dependence. Preliminary results from such a modeling program will be described and compared to empirical radiation belt models.
SM33B-06 14:55h
Electron Transport in the Earth's Outer and Inner Magnetosphere
As electrons are transported from the solar wind and through the Earth's magnetosphere, they can be accelerated to energies $>$ 10 keV when they reach the so-called seed region located at about 10 $R_E$ radially from the Earth in the equatorial plane. As these seed electrons move closer to the Earth, wave-particle interactions can cause further acceleration of electrons to relativistic (MeV) energies. This process occurs during the recovery of magnetic storms, where relativistic electron fluxes are usually observed to be enhanced over pre-storm values in the inner magnetosphere near geosynchronous orbit. In this study, the transport of electrons from the solar wind and outer magnetosphere towards the seed region is examined. This is done by following electron trajectories from different starting points in a global model for the magnetospheric magnetic and electric fields. The electron particle trajectories are followed based on the guiding center approximation and both an empirical and an MHD model (BATS-R-US code) are used for the global magnetospheric fields. In regions where the local fields are very weak (e.g., near reconnection regions), non-adiabatic effects could be important and this is included when following electrons by switching from the guiding center approximation to a full trajectory calculation using the Lorentz force equation in these localized regions of space. Electron distribution functions formed at different locations in the magnetotail as well as in the seed region will be discussed.
SM33B-07 15:10h
The Role Of Non-Adiabatic Processes In The Creation Of The Outer Radiation Belts
The dynamic variation of the electrons in the outer radiation belts has been observed for many years, but the cause of this variation has not been clearly understood. The adiabatic effect due to ring current evolution is not sufficient to account for flux changes during storm phases. Thus, addition processes have been invoked to explain these variations, chiefly the radial diffusion of electrons conserving the first and second adiabatic invariants, where the electrons are accelerated through the betatron and fermi processes, and perhaps with enhanced transport enabled by ULF waves. More recently, local heating inside the radiation belts has been proposed to explain the energization of electrons. However, the exact mechanism that produces the lower energy seed population and converts them to energetic radiation belt electrons has not yet been clarified. In this presentation we will examine the current understanding of the processes which accelerate electrons into the radiation belts with particular emphasis on the role of non-adiabatic processes. The energies that electrons can acquire via radial diffusion are determined using data from ISEE and CRRES, and samples of spectra are examined at various L-values to identify the deficiencies in phase space density. This will determine the extent to which non-adiabatic processes are required to achieve the observed radiation belt energies. We have found strong evidence supporting recent suggestions that non-adiabatic (first invariant breaking) processes are required to explain the generation of the outer belt during intense storm events. We will report on the relative importance of non-adiabatic processes during more typical outer belt conditions.
SM33B-08 15:25h
Compressional Wave Power in the Magnetosphere during Periods of Enhanced Radial Diffusion in the Inner Magnetosphere
Compression fluctuations of the magnetospheric magnetic field travel across magnetic field lines with little attenuation. Thus when the solar wind dynamic pressure is highly fluctuating, the magnetosphere fills with magnetic fluctuations. Examining the Polar magnetic records during the October/November 2003 storms we find that power levels in the frequency bands that resonate with the drift motion of inner zone relativistic electrons increased up to nearly 2000 times the power at quiet times. This activity correlated with enhanced radial diffusion and the formation of a new inner zone electron belt at energies above 1 MeV. Polar magnetic data are also examined at other times when lower rates of diffusion were observed.