Transcript How well can they be applied for space weather modeling

Ring current models:
How well can they be applied
for space weather modeling purposes?
N. Yu. Ganushkina, T. I. Pulkkinen
(Finnish Meteorological Institute, Space Research, Helsinki, Finland),
A. Milillo
(Istituto di Fisica dello Spazio Interplanetario, Rome, Italy),
M. Liemohn
(University of Michigan, Space Physics Research Laboratory, USA)
J. Geophys. Res., 111, A11S08, doi:10.1029/2006JA011609, 2006
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Outline
 Following the evolution of proton ring current during the GEM IM/S Challenge
storm event on April 21-25, 2001 using three ring current models:
- the ring current model combined with tracing particles numerically in the drift
approximation by Ganushkina et al., Ann. Geophys., 23, 579-591, 2005;
- the empirical model of proton fluxes in the inner magnetosphere
by Milillo et al., J. Geophys. Res., 108, doi:10.1029/2002JA009581, 2003;
- the kinetic ring current-atmosphere interaction model (RAM)
by Liemohn et al., J. Geophys. Res., 106, 10,883-10,904, 2001.
 Focus on contributions from protons in different energy ranges to the ring current
energy during different storm phases (base: Polar CAMMICE/MICS observations).
 Study the influence on the model ring current energy of the
- choice of magnetic and electric field models,
- initial particle distributions,
- role of substorm-associated electric fields in particle transport and energization.
 Discussion on outputs from different models.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Vx, km/s IMF Bz, nT
April 21-25, 2001: Storm event overview
10
(a)
ACE
(b)
0
-10
Psw, nPa
-100
-200
-300
-400
-500
12
ACE
(c)
8
4
0
1600
1200
800
400
0
6
4
2
0
40
0
-40
-80
-120
(d)
(e)
SYM-H, nT
Kp
AE, nT
ACE
(f)
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
 Shock arrival: small Vx junp of 100 km/s,
large density pulse with Psw to 10 nPa,
IMF Bz > 0 at 2145 UT Apr 21,
SYM-H from 0 to 30 nT, no AE increase,
Kp around 3
 Cloud arrival: IMF Bz < 0 at 0130 UT Apr 22,
Vx from 400 km/s to 300 km/s,
Psw fluctuating at 3 nPa, SYM-H negative
reaching -100 nT at 1540 UT Apr 22,
AE about 450 nT at 0200 UT Apr 22,
more activation of 1500 nT with SYM-H
decrease, Kp =6 at SYM-H min
 Storm recovery: several SYM-H enhancements
during cloud passage, AE low, Kp below 2,
SW and IMF nominal,
SYM-H to 0 after cloud passage
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ring current energy density and total energy calculated
from Polar CAMMICE/MICS measurements
 Energy density of RC protons

wL   2π 2mq  dE E j E, L , j(E, L) –
0
measured flux.
 Difference for April 21-25, 2001 storm:
Smaller values, dominance of 1-30 keV.
- Polar orbit evolution, ring current fast crossings
at high magnetic latitudes,
- underestimate of 80-200 keV
1.2
0.8
Polar CAMMICE/MICS
 1-200 keV (a)
1-30 keV
 30-80 keV
 80-200 keV
0.4
0
40
SYM-H, nT
 Previous statistical results:
- Before storm: main contribution from 80-200 keV,
- Main phase: dominance of 30-80 keV,
- Recovery: dominance of 80-200 keV.
RC energy, 1014 J
 Total proton ring current energy:
23
3
WRC   w L dV, integrated over RC volume of V=10 m (torus with crossection
of 2.5 Re, radius of 5 Re), symmetric, no PA corrections
V
(b)
0
-40
-80
-120
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model description
 Drift of protons with 90º±60ºpitch angles, 1st and 2nd invariants = const
in time-dependent magnetic and electric fields.
 kappa-type initial distribution with the observed parameters (T, n) by LANL MPA
at R=7 1900-0500 MLT in the equatorial plane.
 Drift velocity as sum of ExB and magnetic (gradient and curvature) drifts:
E B
2p
1/ 2
v  0 0
I  e , I  S 1  Bs  / B  ds,
S
m
0
0
q B
where E0 and B0 are electric
B2
b
0
0
and magnetic fields, p is the
'
m
m
particle moment, q is the particle charge, τb is the bounce period, Bm is the magnetic
field and mirror point Sm, ds is the element of magnetic field line length.
 Changes in distribution function and flux calculations using Liouville’s theorem
(conservation of distribution function along dynamic trajectory of particles)
taking into account charge-exchange processes with cross section by
Janev and Smith, 1993 and number density of neutrals by thermosphere model
MSISE 90 (Hedin, 1991)
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model:
Initial boundary conditions for April 21-25, 2001 storm
 kappa-type initial distribution with n and TII and T by LANL MPA;
 Time-dependent boundary conditions:
- measurements within 4 h around midnight;
- values averaged for more than one satellite
simultaneously in the region;
- linear interpolation of data when no satellite.
T II, T  , keV
n is the particle number density,
m is the particle mass,
E0 = kB T(1-3/2k) is the distribution peak
particle energy, kB = 1.3807*10-23 J/K is the
Boltzmann constant, T = 1/3 (TII + 2T),
gamma functions computed for k=5.
n, cm-3
3/ 2
  k  1
 m 
k  1 
E 


f E   n
1
,
 2kE 


k  1 / 2
kE

0

0
2.4
2
1.6
1.2
0.8
0.4
16
12
where
LANL MPA, nightside
(a)
(b)
T
8
4
TII
0
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model:
Models for electric fields for April 21-25, 2001 storm
Electric field models:
(1) Kp-dependent Volland-Stern convection electric field with observed Kp
 convection  AR sin   0 ,
  2, A 
0.045
2
kV
/
R
E
1  0.159 Kp  0.0093Kp 2
 is the magnetic local time, 0 = 0 is the offset angle from dawn-dusk meridian;
(2) Boyle et al., 1997 polar cap potential dependent on solar wind and IMF
parameters applied to Volland-Stern convection field

4 2
3   IMF
  1.1 10 Vsw  11.1BIMF sin 
 2

RB  10.47 RE
2
 sin   R 
  ,

 2  RB 
IMF is the IMF clock angle.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model:
Models for magnetic fields for April 21-25, 2001 storm
Magnetic field models:
(1) dipole;
(2) Tsyganenko T89 model parameterized by Kp, Kp observed;
(3) Tsyganenko T01s model with the observed input parameters such as Dst index,
solar wind dynamic pressure Psw, IMF By and IMF Bz, functions G1 and
G2, which depend on IMF Bz and Vsw and take into account the history of
solar wind.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model results for April 21-25, 2001 storm :
Influence of electric and magnetic field models
1-200 keV
1-30 keV
4
dip+VS (a)
3
2
2
1
0
4
3
dip+T89+VS
2
1
0
4
dip+T01s+VS
3
2
Proton ring current energy, 1014 J
3
0
4
3
0
4
3
-120
dip+T01s+Boyle
2
0
40
-80
dip+T89+Boyle
1
0
40
-40
No initial distribution
in the inner
magnetosphere
2
1
0
dip+Boyle (b)
1
1
SYM-H , nT
SYM-H , nT
Proton ring current energy, 1014 J
4
30-80 keV
80-200 keV
0
-40
-80
-120
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model: Electric field pulse model
Time varying fields associated with dipolarization in magnetotail, modeled as
an electromagnetic pulse (Li et al., 1998; Sarris et al., 2002):
 Perturbed fields propagate from tail toward the Earth;
 Time-dependent Gaussian pulse with azimuthal E;
 E propagates radially inward at a decreasing velocity;
 decreases away from midnight.
Time-dependent B from the pulse is calculated by Faraday’s law.
In spherical coordinates (r, , ):
  r  ri  vr t  t a / d
v r   a  br
 
E   E0 1  c1 cos  0 p exp  2 ,
- location of the pulse maximum,
r I determines pulse arrival time
- pulse front velocity, d - width of pulse,
c1 , p describe LT dependence of E amplitude, largest at 0,
t a  c2 R E / v a 1  cos  0 
- delay of pulse from 0 to other LTs,
c2 - delay magnitude,
va - longitudinal propagation speed
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al. model: Addition of pulsed
electromagnetic field for April 21-25, 2001 storm
0435
2.8
1
0805 1000
3.0 4.4
2 3
1420 1730
6.0
4.0
1945
1325 1540
5.6
4.8
4.0
8
4 5 6 7
0630
2.8
0130
4.4
9
10
1800
1500
AE, nT
1200
900
600
300
0
0
3
6
9
15
12
April 22
18
21
0
3
6
9
15
12
April 23
18
UT
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
21
24
Ganushkina et al model results for April 21-25, 2001 storm :
Role of smaller-scale electric fields
30-80 keV
80-200 keV
1-200 keV
1-30 keV
4
dip+VS+pulses (a)
3
2
2
1
0
4
dip+T89+VS+pulses
3
2
1
0
4
dip+T01s+VS+pulses
3
2
Proton ring current energy, 1014 J
3
0
4
2
1
0
4
2
0
40
-80
-120
dip+T01s+Boyle+pulses
3
0
40
-40
dip+T89+Boyle+pulses
3
1
0
dip+Boyle+pulses (b)
No initial distribution
in the inner
magnetosphere
1
1
SYM-H , nT
SYM-H , nT
Proton ring current energy, 1014 J
4
0
-40
-80
-120
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model:
Initial distribution in the inner magnetosphere
for April 21-23, 2001 storm
 Available observational data not sufficient
to reconstruct the prestorm initial proton
distribution in the inner magnetosphere.
Initial energy density
distribution
 As initial distribution - model distribution
obtained at the end of Apr 25, 2001 after
tracing with empty inner magnetosphere.
Final ED map while tracing
with empty magnetosphere
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model results for April 21-25, 2001 storm :
Effects on the initial distribution
1-200 keV
1-30 keV
4
dip+VS (a)
3
2
2
1
0
4
3
dip+T89+VS
2
1
0
4
dip+T01s+VS
3
2
Proton ring current energy, 1014 J
3
0
4
2
1
0
4
2
0
40
-80
-120
dip+T01s+Boyle
3
0
40
-40
dip+T89+Boyle
3
1
0
dip+Boyle (b)
Initial distribution
in the inner
magnetosphere
1
1
SYM-H , nT
SYM-H , nT
Proton ring current energy, 1014 J
4
30-80 keV
80-200 keV
0
-40
-80
-120
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Ganushkina et al model results for April 21-25, 2001 storm :
Effects on the initial distribution and smaller-scale fields
1-200 keV
1-30 keV
4
dip+VS+pulses (a)
3
2
2
1
0
4
dip+T89+VS+pulses
3
2
1
0
4
dip+T01s+VS+pulses
3
2
Proton ring current energy, 1014 J
3
0
4
2
1
0
4
2
0
40
-80
dip+T01s+Boyle+pulses
3
0
40
-40
dip+T89+Boyle+pulses
3
1
0
dip+Boyle+pulses (b)
Initial distribution
in the inner
magnetosphere
1
1
SYM-H , nT
SYM-H , nT
Proton ring current energy, 1014 J
4
30-80 keV
80-200 keV
0
-40
-80
-120
-120
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Milillo et al. model description
 Based on AMPTE/CHEM data of 90 PA H+ fluxes at L=3-9.3 for 1.5-316 keV.
 Gives ion distributions as a function of L, energy, MLT.
 Functional form of model distribution consists of:
(1) Gaussian in L added to a continuum with a Gaussian shape in energy for
proton flux at intermediate energies of 5-80 keV –
CONVECTION/INJECTION POPULATION
(2) Gaussian for high energy population (> 40 keV) –
DIFFUSION POPULATION
 For April 21-25, 2001 storm LANL MPA data for 3-45 keV and
SOPA data for 50-400 keV added.
 Set of 6 time-evolving parameters (intensity, energy position and width of two
populations) applied to the model, model gives storm evolution of two populations.
 Total energy is calculated by integrating energy density.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Liemohn et al. model description
 Kinetic ring current-atmosphere interaction model (RAM)
- solves the gyration and bounce-averaged Boltzmann equation inside of
geostationary orbit;
- uses second-order accurate numerical schemes to determine hot ion phase
space distribution as a function of time, equatorial plane location, energy,
and equatorial pitch angle.
 Initial conditions from Sheldon and Hamilton [1993].
 Sources specified by LANL MPA and SOPA data across the nightside
outer boundary.
 Loss mechanisms include
- flow of plasma out the dayside outer boundary,
- precipitation of particles into the upper atmosphere,
- pitch angle scattering and drag from Coulomb collisions (plasmaspheric
model of Ober et al. [1997]),
- charge exchange with neutral hydrogen geocorona (Rairden et al. [1986]).
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Milillo et al., Liemohn et al. and Ganushkina et al.
model results for April 21-25, 2001 storm : Comparison
1-200 keV
1-30 keV
30-80 keV
80-200 keV
8
4
0
16
total
12
3-200 keV
3-30 keV
30-80 keV
(b)
80-200 keV
16
4
0
16
convected
(c)
12
8
4
0
16
diffused
(d)
12
8
Ring current energy, 1014 J
8
4
(e)
0
-40
-80
dip + VS (a)
12
8
4
0
16
12
dip + self-cons E (b)
8
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
2
1
0
4
2
1
0
4
2
0
40
-40
-80
dip+T01s+VS+pulses
3
0
40
0
dip+T89+VS+pulses
3
1
(c)
dip+VS+pulses
3
4
0
-40
-80
-120
-120
-120
4
Proton ring current energy, 1014 J
(a)
SYM-H , nT
12
0
40
SYM-H, nT
total
convected
diffused
SYM-H , nT
Ring current energy, 1014 J
Ring current energy, 1014 J
16
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
0 12 0 12 0 12 0 12 0 12 24
Apr 21 Apr 22 Apr 23 Apr 24 Apr 25
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Summary
Modelling of April 21-25, 2001 storm event by three ring current models
All models predict
- high-energy protons dominate before the storm,
- the low and medium energy protons are rapidly enhanced during the main
phase of the storm,
- slowly decline in energy content throughout the recovery phase of the storm.
- main difference between the models is in the contribution from the high-energy
protons during the storm.
 Ganushkina model predicts a low contribution from these protons,
unless an extra electric field is included to replicate substorm injections.
 The Milillo model predicts that the high-energy protons dominate throughout
the storm.
 The Liemohn model predicts a constant contribution from the high-energy
protons during the late recovery phase.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
General Summary
 Changing from dipole to more realistic magnetic field decreased the RC energy
content by 30%.
 Details and strength of convection electric field cause only small changes in
time-evolving RC energy content.
 Time-dependent and localized electric fields are the only means to provide
preferential increase of high-energy particles.
 Role of diffused processes is rather small in bringing RC ion during main phase.
 Relative contributions from diffusion and convection to RC energy content are
equal during recovery phase.
 Initial populations in the inner magnetosphere and boundary conditions have
significant effects on model results.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Summary: Model results to Dst
Converting peak energy values to magnetic perturbations at the Earth’s surface
DPS formulation:
B = 3.98 * 10-30 ERC
Ganushkina model, ERC = 4*1014 J, B = 10 nT
Milillo and Liemohn models, ERC = 14*1014 J, B = 35 nT
Observations: 100 nT
- Models significantly underestimate the total energy in the ring current region
- DPS relation does not account for contributions of other current systems to Dst
Include of acceleration processes into models, which are significant in
producing high-energy populations (can affect humans and/or technological
systems in space)
Need for realistic electric field models!
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
References (1)
Boyle, C. B., P. H. Reiff, and M. R. Hairston (1997), Empirical polar cap potentials,
J. Geophys. Res., 102, 111-125.
Ganushkina, N. Yu., T. I. Pulkkinen, T. Fritz (2005), Role of substorm-associated
impulsive electric fields in the ring current development during storms, Ann.
Geophys., 23, 579-591.
Hedin, A. E. (1991), Extension of the MSIS thermosphere model into the middle and
lower atmosphere, J. Geophys. Res., 96, 1159.
Janev, R. K. and J. J. Smith (1993), Cross sections for collision processes of
hydrogen atoms with electrons, protons, and multiply-charged ions, in: Atomic and
Plasma-Material Interaction Data for Fusion, Int. At. Energ. Agency, 4.
Li, X., D. N. Baker, M. Temerin, et al. (1998), Simulation of dispersionless
injections and drift echoes of energetic electrons associated with substorms,
Geophys. Res. Lett., 25, 3763-3766.
Liemohn, M. W., J. U. Kozyra, M. F. Thomsen, et al., Dominant role of the
asymmetric ring current in producing the stormtime Dst (2001), J. Geophys. Res.,
106, 10,883-10,904.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
References (2)
Milillo A., S. Orsini, and I. A. Daglis (2001), Empirical model of proton fluxes in the
equatorial inner magnetosphere. 1. Development, J. Geophys. Res., 106, 2571325730.
Milillo A., S. Orsini, D. C. Delcourt, A. Mura, S. Massetti, E. De Angelis, and Y.
Ebihara (2003), Empirical model of proton fluxes in the equatorial inner
magnetosphere: 2. Properties and applications, J. Geophys. Res., 108,
doi:10.1029/2002JA009581.
Ober, D. M., J. L. Horwitz, and D. L. Gallagher (1997), Formation of density troughs
embedded in the outer plasmasphere by subauroral ion drift events, J. Geophys. Res.,
102, 14,595.
Orsini, S., A. Milillo, A. Mura (2004), modeling of the Inner Magnetospheric TimeEvolving Plasma: an empirical approach based on proton distribution, J. Geophys.
Res., 109, A11216, doi: 10.1029/ 2004JA010532.
Rairden, R. L., L. A. Frank, and J. D. Craven (1986), Geocoronal imaging with
Dynamics Explorer, J. Geophys. Res., 91, 13,613.
Sarris, T. E, X. Li, N. Tsaggas, and N. Paschalidis (2002), Modeling energetic
particle injections in dynamic pulse fields with varying propagation speeds, J.
Geophys. Res., 107, 1033, doi:10.1029/2001JA900166.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
References (3)
Sheldon R. B., and D. C. Hamilton (1993), Ion transport and loss in the Earth's quiet
ring current 1. Data and standard model, J. Geophys. Res., 98, 13491-13508.
Stern, D. P. (1975), The motion of a proton in the equatorial magnetosphere, J.
Geophys. Res., 80, 595-599.
Tsyganenko, N. A. (1989), A magnetospheric magnetic field model with a warped
tail current sheet, Planet. Space Sci., 37, 5-20.
Tsyganenko, N. A. (2002a), A model of the near magnetosphere with a dawn-dusk
asymmetry: 1. Mathematical structure, J. Geophys. Res., 107, 1179,
doi:10.1029/2001JA0002192001.
Tsyganenko, N. A. (2002b), A model of the near magnetosphere with a dawn-dusk
asymmetry: 2. Parameterization and fitting to observations, J. Geophys. Res., 107,
1176, doi:10.1029/2001JA000220.
Volland, H. (1973), A semi-empirical model of large-scale magnetospheric electric
field, J. Geophys. Res., 78, 171.
Wilken, B. et al. (1992), Magnetospheric ion composition spectrometer onboard the
CRRES spacecraft, J. Spacecraft and Rockets, 29, 585.
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium
Contributions to RC energy from protons
with different energy ranges: 27 storms’ statistics
20-80 keV
1-20 keV
initial phase
main phase
100
100
80
80
60
60
60
40
40
40
20
0
100
80
60
40
20
0
100
80
contribution to total energy range, %
80
contribution to total energy range, %
contribution to total energy range, %
100
80-200 keV
20
0
100
80
60
40
20
100
80
20
0
100
80
60
40
20
0
100
80
60
60
60
40
40
40
20
20
20
0
0
0
60 40 20 0 -20 -40
Dst, nT
0
-50 -100 -150 -200
Dst, nT
recovery phase
-100
-50
0
Dst, nT
50
3d European Space Weather Week – November 13-17, 2006, Brussels, Belgium