New Insights on Geomagnetic Storms From Model Simulations Using Multi-Spacecraft Data : • Solar and interplanetary origin of geomagnetic storms
• Sources, acceleration, and losses of ring current ions
• Modeling the evolution of the terrestrial ring current using multi-satellite data by
Vania K. Jordanova
Space Science Center/EOS
Department of Physics
University of New Hampshire, Durham, USA New Insights on Geomagnetic Storms From Model Simulations Using Multi-Spacecraft Data
Geomagnetic Storm: Ring Current Evolution : main recovery phase Sudden
Commencement Geomagnetic Storm: Ring Current Evolution
Geomagnetic Storm: Ring Current Evolution : • Composition: e-, H+, He+, O+, N+, He++
• Energy Range: ~ 1 keV < E < 300 keV
• Location: ~ 2 < L < 8
• Energy Density: ~ 10 - 1000 keV/cm3 main recovery phase Sudden
Commencement Geomagnetic Storm: Ring Current Evolution
Solar - Interplanetary - Magnetosphere Coupling : Solar - Interplanetary - Magnetosphere Coupling • Flow of plasma within the magnetosphere (convection) [Gonzalez et al., 1994]
Solar - Interplanetary - Magnetosphere Coupling : Solar - Interplanetary - Magnetosphere Coupling • Flow of plasma within the magnetosphere (convection) [Gonzalez et al., 1994]
Sources of Ring Current Ions : Sources of Ring Current Ions [Chappell et al., 1987] • Solar wind
• Ionosphere
Sources of Ring Current Ions : Sources of Ring Current Ions [Chappell et al., 1987] max H+: solar min & quiet conditions
max O+: solar max & active conditions
Total ionospheric flux ~ 10 26 ions/s
=> comparable to solar wind source • Solar wind
• Ionosphere
Ring Current Loss Processes : Ring Current Belt
(1-300 keV)
Density Isocontours
Dawn Dusk Lower Density Cold
Plasmaspheric Plasma
(Dusk Bulge Region) ( L~6 ) ( L~8 ) Plasmapause ( L~4) [Kozyra & Nagy, 1991] Ring Current Loss Processes
Ring Current Loss Processes : Ring Current Belt
(1-300 keV)
Density Isocontours
Dawn Dusk Energetic
Neutral
Precipitation Lower Density Cold
Plasmaspheric Plasma
(Dusk Bulge Region) ( L~6 ) ( L~8 ) Plasmapause ( L~4) Charge
Exchange [Kozyra & Nagy, 1991] Ring Current Loss Processes
Ring Current Loss Processes : Ring Current Belt
(1-300 keV)
Density Isocontours
Dawn Dusk Conjugate
SAR Arcs Energetic
Neutral
Precipitation Anisotropic
Energetic
Ion Precipitation Coulomb
Collisions
Between
Ring Currents
and
Thermals
(Shaded Area) Lower Density Cold
Plasmaspheric Plasma
(Dusk Bulge Region) ( L~6 ) ( L~8 ) Plasmapause ( L~4) Charge
Exchange [Kozyra & Nagy, 1991] Ring Current Loss Processes
Ring Current Loss Processes : Ring Current Belt
(1-300 keV)
Density Isocontours
Dawn Dusk Conjugate
SAR Arcs Energetic
Neutral
Precipitation Anisotropic
Energetic
Ion Precipitation Coulomb
Collisions
Between
Ring Currents
and
Thermals
(Shaded Area) Lower Density Cold
Plasmaspheric Plasma
(Dusk Bulge Region) ( L~6 ) ( L~8 ) Wave Scattering
of Ring Current Ions Plasmapause ( L~4) Isotropic Energetic Ion
Precipitation
Ion
Cyclotron
Waves Charge
Exchange [Kozyra & Nagy, 1991] Ring Current Loss Processes
Theoretical Approaches : • Single particle motion - describes the motion of a particle under the influence of external electric and magnetic fields
• Magnetohydrodynamics and Multi-Fluid theory - the plasma is treated as conducting fluids with macroscopic variables
• Kinetic theory - adopts a statistical approach and looks at the development of the distribution function for a system of particles Theoretical Approaches
Kinetic Model of the Terrestrial Ring Current : - radial distance in the equatorial plane from 2 to 6.5 RE
- azimuthal angle from 0 to 360
- kinetic energy from 100 eV to 400 keV
- equatorial pitch angle form 0 to 90
- bounce-averaging (between mirror points) and where Kinetic Model of the Terrestrial Ring Current • Initial conditions: POLAR and EQUATOR-S data
• Boundary conditions: LANL/MPA and SOPA data [Jordanova et al., 1994; 1997]
Charge Exchange Model : Equatorial exospheric Hydrogen densities
[Rairden et al., 1986] Charge exchange cross sections
[Phaneuf et al., 1987; Barnett, 1990] Charge Exchange Model
Plasmasphere Model : Plasmasphere Model Equatorial plasmaspheric electron density
Ion composition: 77% H+, 20% He+, 3% O+
Plasmasphere Model : Plasmasphere Model Equatorial plasmaspheric electron density
Ion composition: 77% H+, 20% He+, 3% O+ Comparison with geosynchronous LANL data
Wave-Particle Interactions Model : Wave-Particle Interactions Model
where nt, EII, At are calculated with our kinetic model for H+, He+, and O+ ions
Integrate the local growth rate along wave paths and obtain the wave gain G(dB) Solve the hot plasma dispersion relation for
EMIC waves:
Wave-Particle Interactions Model : Wave-Particle Interactions Model Solve the hot plasma dispersion relation for EMIC waves:
where nt, EII, At are calculated with our kinetic model for H+, He+, and O+ ions
Integrate the local growth rate along wave paths and obtain the wave gain G(dB)
Use a semi-empirical model to relate G to the wave amplitude Bw:
[Jordanova et al., 2001]
Wave-Particle Interactions Model : Wave-Particle Interactions Model Solve the hot plasma dispersion relation for EMIC waves:
where nt, EII, At are calculated with our kinetic model for H+, He+, and O+ ions
Integrate the local growth rate along wave paths and obtain the wave gain G(dB)
Use a semi-empirical model to relate G to the wave amplitude Bw:
[Jordanova et al., 2001]
WIND Data & Geomagnetic Indices : WIND Data & Geomagnetic Indices • Magnetic cloud
• Moderate geomagnetic storm Dst=-83 nT & Kp=6
Model Results: Dst Index, Jan 10, 1997 : Model Results: Dst Index, Jan 10, 1997 Comparison of:
• Kp-dependent Volland-Stern model
• IMF-dependent Weimer model
=> Weimer model predicts larger electric field, which results in larger injection rate and stronger ring current buildup
Effects of Wave-Particle Interactions : Model results & HYDRA data comparison:
• Pitch angle scattering has larger effect than energy diffusion
• Non-local effects of WPI due to transport Effects of Wave-Particle Interactions
Effects of Collisional Losses : Effects of Collisional Losses Comparison of model results with POLAR data Larger effect on:
- postnoon spectra
- low L shells
- high magnetic latitudes
- slowly drifting ~1-30 keV ions
Effects of Time-Dependent Plasmasheet Source Population: October 1995 : • Enhancement in the convection electric field alone is not sufficient to reproduce the stormtime Dst
• The strength of the ring current doubles when the stormtime enhancement of plasmasheet density is considered Effects of Time-Dependent Plasmasheet Source Population: October 1995
Effects of Inner Magnetospheric Convection: March 9-13, 1998 : Effects of Inner Magnetospheric Convection: March 9-13, 1998 Electric potential in the equatorial plane:
• Both models predict strongest fields during the main phase of the storm
• Volland-Stern model is symmetric about dawn/dusk by definition
• Weimer model is more complex and exhibits variable east-west symmetry and spatial irregularities
Modeled H+ Distribution and POLAR Data: March 1998 : Modeled H+ Distribution and POLAR Data: March 1998 HYDRA Volland-Stern Model Weimer Model
Bounce-averaged Drift Paths of Ring Current Ions : • East-West transition occurs at lower energy in Volland-Stern model
• Particles follow drift paths at larger distances from Earth and experience less collisional losses in Weimer model Bounce-averaged Drift Paths of Ring Current Ions
Ring Current Energization & Dst: July 13-18, 2000 : Ring Current Energization & Dst: July 13-18, 2000
Ring Current Asymmetry & Ion Composition : • A very asymmetric ring current distribution during the main and early recovery phases of the great storm
• Near Dst minimum O+ becomes the dominant ion in agreement with previous observations of great storms Ring Current Asymmetry & Ion Composition
EMIC Waves Excitation : • Intense EMIC waves from the O+ band are excited near Dst minimum
• The wave gain of the O+ band exceeds the magnitude of the He+ band
• EMIC waves from the O+ band are excited at larger L shells than the He+ band waves EMIC Waves Excitation
Ion Pitch Angle Distributions from POLAR/IPS : • Data are from the northern pass at ~hour 75 (left) and from the southern pass at ~hour 93 (right), MLT~16
• Isotropic pitch angle distributions, indicating strong diffusion scattering are observed at large L shells near Dst minimum
• Partially filled loss cones, indicating moderate diffusion are observed at lower L shells and during the recovery phase Ion Pitch Angle Distributions from POLAR/IPS L=7 L=6 L=5 L=4 L=3
Model Results: Precipitating Proton Flux : Model Results: Precipitating Proton Flux Hour 75
Model Results: Precipitating Proton Flux : • Precipitating H+ fluxes are significantly enhanced by wave-particle interactions
• Their temporal and spatial evolution is in good agreement with POLAR/IPS data Model Results: Precipitating Proton Flux Hour 75 Hour 93
Proton Ring Current Energy Losses : • Proton precipitation losses increase by more than an order of magnitude when WPI are considered
• Losses due to charge exchange are, however, predominant Proton Ring Current Energy Losses
Conclusions : The ring current is a very dynamic region that couples the magnetosphere and the ionosphere during geomagnetic storms
New results emerging from recent simulation studies were discussed:
• the effect of the convection electric field on ring current dynamics: influence on Dst index, east-west transition energy, dips in the distribution function
• the important role of the stormtime plasmasheet enhancement for ring current buildup
• the formation of an asymmetric ring current during the main and early recovery storm phases
• it was shown that charge exchange is the dominant ring current loss process
• wave-particle interactions contribute significantly to ion precipitation, however, their effect on the total energy balance of the ring current population is only ~2-8% reduction
More studies are needed
• to determine the effect of WPI on the heavy ion components, moreover O+ is the dominant ring current specie during great storms
• to determine the contribution of substorm induced electric fields on ring current dynamics
Conclusions
Acknowledgments : Many thanks are due to:
C. Farrugia, L. Kistler, M. Popecki, and R. Torbert,
Space Science Center/EOS, University of New Hampshire, Durham
R. Thorne, Department of Atmospheric Sciences, UCLA, CA
J. Fennell and J. Roeder, Aerospace Corporation, Los Angeles, CA
M. Thomsen, J. Borovsky, and G. Reeves, Los Alamos Nat Laboratory, NM
J. Foster, MIT Haystack Observatory, Westford, MA
R. Erlandson, Johns Hopkins University, APL, Laurel, MD
K. Mursula, University of Oulu, Oulu, Finland
This research has been supported in part by NASA under grants NAG5-7804, NAG5-4680, NAG5-8041 and NSF under grant ATM 0101095 Acknowledgments