Sinaia, September 6-10, 20051 Berndt Klecker Max-Planck-Institut für extraterrestrische Physik, 85741 Garching, Germany Workshop on Solar Terrestrial Interactions.

Sinaia, September 6-10, 2005 1

Berndt Klecker

Max-Planck-Institut für extraterrestrische Physik, 85741 Garching, Germany

Workshop on

Solar Terrestrial Interactions from

Microscale to Global Models

Sinaia, Romania, September 6 - 10, 2005

Heavy Ion Charge States in

Solar Energetic Particle Events


• Introduction

• Measurement Techniques

• Ionic Charge State (Fe , Ne, Mg, Si) in IP Shock /CME Related SEP

Events

• Ionic Charge States (Fe, Ne, Mg, Si) in 3He-rich and Heavy Ion-rich Events

• The Energy Dependence of Ionic Charge States - Mechanisms

• Summary

OUTLINE


ENEGETIC PARTICLES IN THE HELIOSPHERE


• Information on the Source

i.e. Solar (Solar Wind, Corona); Interstellar, e.g. He+ Pickup Ions

For Solar Source: Source Location (Temperature, Density)

• Important Information on Fractionation, Acceleration and Propagation

Processes

Injection, Acceleration and Propagation generally depend on Rigidity,

i.e. particle velocity v and M/Q

INTRODUCTION

Why are Ionic Charge States Important?


WHERE DO SOLAR ENERGETIC PARTICLES COME FROM ?The Historical Development

Forbush, 1946

Phase 1:

Everything comes from Flares

Phase 2: ~ 70s to 90s

Flares and CMEs / Shocks

Impulsive and Gradual SEPs

Phase 3: Present

Flares and CMEs / ShocksRelative Contribution to SEPs under Debate

Classification of 2 distinct types of SEPs events in question.

IMPULSIVEFLARES

GRADUALFLARES

DurationSXR < 1 h < 10 hγ SXR < 10 min < 10 min

Height ≤10 km ~5 ⋅10

Volume 10 -10 cm 10 -10 cmenerg y density high LowHa size small LargeDuration HXR < 10 min > 10 minDuration m < 5 min > 5 minMetri c Radio (I )I , III II,(III),IV

Lin, 1970; Pallavicini et al., 1977, Reames 1999

He-rich gradualparticles electron rich proton rich

He/ He ~ 1 ~0.0005Fe/O ~ 1.23 ~ 0.15H/He ~ 10 ~ 100Q ~ 20 ~ 14Duration hours Days

Long. Distrib < 30° ≤ 80°Metric Radio III, V II,III,IV,VSolar Wind - Ipl. shockEvent Rate ~ 1000/a ~ 10/a

1st measurement of 2 GLEs

in 1942


Average of 20 Events

Energy: 385 keV/nuc

IMPULSIVE EVENTS Average Elemental Abundances

• Mason et al., 2002, 2004

• Reames, 1999

NEW


Results from early measurements at ~1 Mev/nuc:

Qm(Fe) ~12 -16

-> Te ~ 1.5-2 106 K

Coronal Temperatures

Q ~ Solar Wind, but somewhat larger (Fe)

EARLY RESULTS

for Large (gradual) IP-Shock Related SEP Events

Gloeckler et. al., 1976, Hovestadt et al. 1981, Luhn et al., 1984


Qm (Fe) ~ 19-20, Qm (Si) ~14

-> Te ~107 K

EARLY RESULTS

for 3He-, Fe-rich (Impulsive) SEP Events

Klecker et al., 1984, Luhn et al., 1987


EARLY RESULTS

Puzzle:

Gradual: Q at ~1 MeV/n similar to Solar Wind, but for some

ions (e.g. Fe) higher than in Solar Wind

Impulsive: Si fully ionized, i.e. M/Q=2

How can abundances be enhanced relative to C or O

(M/Q=2 for C - Si)

Question: Measurement only in small energy range at

~1MeV/nuc. How is Q at other energies?


1) In-Situ Measurement (e.g. by Electrostatic Deflection)

Energy range from Solar Wind energies to a few MeV/amu

Advantage: Direct Measurement of E, M, Q, Q Distribution,

Energy Dependence Q (E)

2) Measurement of the Rigidity Cutoff in the Earth’s Magnetic Field

Measurement of M, E, Rcutoff > Determination of average Q

Advantage: Q Determination to High Energies of 10s of MeV/amu

3) Indirect Methods using information on Energy Spectra, Composition, or time-

intensity profiles

Disadvantage: Model dependent

IONIC CHARGE DETERMINATION

Measurement Techniques


We want: E, M, Q Measurement of E/Q (electrostatic defl.)

E/M (e.g. time-of-

flight)

E (SSD)

Solar Wind: SWICS / Ulysses, SWICS/ACE, CTOF/SOHO

Suprathermal: STOF /SOHO, SEPICA/ACE

~ 0.2 - 0.6 Mev/nuc: SEPICA/ACE

~ 0.5 - 2.0 Mev/nuc: IMP-7/8, ISEE-1/3


(1) In-Situ Measurements


1) In-Situ Measurement (e.g. by Electrostatic Deflection)

Energy range from Solar Wind energies to a few MeV/amu

Advantage: Direct Measurement of E, M, Q, Q Distribution,

Energy Dependence Q (E)

2) Measurement of the Rigidity Cutoff in the Earth’s Magnetic Field

Measurement of M, E, Rcutoff > Determination of average Q

Advantage: Q Determination to High Energies of 10s of MeV/amu

3) Indirect Methods using information e.g. on Energy Spectra, Composition, or

time- intensity profiles






(2) Rigidity Cutoff of the Earth’s Magnetic Field

10 1

10 2

55 60 65 70 75

Sampex/Lica0.85-1.25 MeV/nuc

Flux (particles/cm

2

-sec-sr-MeV/n)

Adjusted invariant latitude

normalization:average flux between 75-85degrees invariantlatitude

1H4He

16 O

Fe (group)

97311 0005 - 97313 1156

Mason et al., 1995; Mazur et al., 1995;

Leske et al., 1995; Oetliker et al., 1997

• Determine c(Rc) with ions of known charge (H+) on an orbit-by orbit bases

• Determine c for other ions

• Compute Qavg from Rc, c and E, M

Advantage:

Large Energy Range

Energy Dependence

Disadvantage:

Intensity needs to be large

SAMPEX

(polar Orbit, ~ 600 km altitude)



(2) Rigidity Cutoff Variations During SEP Events

Leske et al., 2001

• c(Rc) can vary by several degrees

during an event• Determine c for H+ or He2+ on an orbit by

orbit basis• Compute adjusted c from time variation

• Use c(Rc) or linear fit: cos4(c) = a Rc+b

to derive Qavg from Rc, c and v, M

Qavg = (M v) / (Rc e)

SAMPEX


3) Indirect Methods using information e.g. on Energy Spectra, Composition, or

time - intensity profiles


• Energy Spectra: M/Q dependent roll-over of spectra (Tylka et al.,

2000)

• Composition: M/Q-dependent fractionation effects (Cohen et al.,

1999)

Rigidity dependent interplanetary propagation:

• Time to maximum intensity (O’Gallagher et al, 1976, Dietrich & Tylka, 2003)

• SEP decay phase (Sollitt et al., 2003)





(3) Indirect Methods

1. FeX(E) ~ Eγ exp(-E/E0X)

2. E0X =E0H*(Q/M) 1

April 20-24, 1998Tylka et al., 2000

• Determine E0X, γ from spectral fit

• Determine M/Q from (2)

17

IONIC CHARGE DETERMINATIONExperiments and Energy Range

EEARLY MEASUREMENTS FROM IMP-7 / 8, ISEE - 1/3

RECENT MEASUREMENTS FROM SAMPEX - SOHO - ACE

0

4

10-1 100 101 102 103 104 105

ENERGY (keV/nucleon)

Solar Wind

SWICS / ACE

Suprathermal and Energetic Particles

STOF / CELIAS / SOHO

SEPICA / ACE

LICA+HILT+MAST / SAMPEX

ULEZEQ / ISEE-1/3


NEW RESULTS (SAMPEX-SOHO-ACE) Gradual Events: Mean Ionic Charge Varies With Energy

SAMPEX: Mason et al., 1995; Leske et al., 1995, Oetliker et al., 1997)

Systematic Increase of Q with Energy above ~10 MeV/amu, in particular for Fe

Oct. 1992


NEW RESULTS (SAMPEX-SOHO-ACE) Gradual Events: Large Variability of Q (E)

Möbius et al., 1999, 2000, 2003; Bogdanov et al., 2000, Klecker et al. 2000, 2001, 2003; Popecki et al., 2000, 2001, 2003; Bamert et al., 2002; Labrador et al., 2003

Large Variability of Q (E) for Heavy Ions, in particular for Fe

At energies above ~200 keV/nuc:

Large VariabilityQFe(E) increasing at E > 10 Mev/nuc - often

QFe (E) increasing at ~ 1 MeV/nuc - some cases

At low energies of up to ~ 250 keV/amu:

Q similar to Solar Wind

0

10

20

30

40

50

60

0 5 10 15 20 25

Q STOF (x3)Q SEPICA (0.18-0.25 MeV/n)

IONIC CHARGE

1998, Day 121

Day 121, 1998 CME / IP Shock Event

0.01 - 0.1 MeV/n

SW: 10.1


NEW RESULTS (SAMPEX-SOHO-ACE) Gradual Events: Mean Ionic Charge Varies With Energy

SAMPEX ResultsMason et al., 1995; Leske et al. 1995; Oetliker et. 1997; Mazur et al., 1999;Leske et al., 2001; Labrador et al., 2003

ACE ResultsMöbius et al., 1999, 2000, 2003; Bogdanov et al., 2000, Klecker et al. 2000, 2001, 2003; Popecki et al., 2000, 2001, 2003


NEW RESULTS (ACE+SOHO) Impulsive Events: Mean Ionic Charge Increases ALWAYS with Energy

YEAR DATE Qm (Fe)0.18-0.43

Q

1998 252 00:29-253 23:45

17.5 0.60

1999 184 21:36-186 06:00

14.9 0.60

1999 201 02:19-202 22:19

16.5 0.60

2000 122 04:05-122 23:54

15.2 0.55

Möbius et al., 2003; Klecker et al, 2005

8

10

12

14

16

18

20

22

24

0.01 0.1

Event 1

Event 2

Event 3

Event 4

STOF-AVG (2-4)

Averge Charge of Fe

Energy (MeV/nuc)


IMPULSIVE EVENTS Ionic Charge of Ne, Mg, Si, Fe (ACE)

6

10

14

18

22

26

0.1 1

9. September 1998

Q-NeQ-MgQ-SiQ-Fe

Energy (MeV/nuc)

SW / CME related SEP

6

10

14

18

22

26

0.1 1

20. July 1999

Q-Ne

Q-Mg

Q-Si

Q-Fe

Energy (MeV/nuc)

SW / CME related SEP


THE ENERGY DEPENDENCE OF THE IONIC CHARGE Overview of Possible Mechanisms

1) Ionization by e, p in a dense plasma in the low corona

“Stripping Model”

2) Effect of Energy Spectra with M/Q-dependent roll-over

(i.e. Acceleration and Propagation effects)

2) Mixing of 2 Sources: Solar Wind Origin and Flare Origin (i.e Heavy Ion Rich)


Comparison of Ionic Charge States with Stripping ModelI. The Equilibrium Case

The Equilibrium Case

Impact ionization by p + e

Radiative + dielectronic recombination

1. Qm at E < 0.1 MeV/amu depends on Te (electron distribution function)

2) Large Increase of Qm at E > 0.1 MeV/n by (p+e) impact ionization

Electrons: Maxwell distribution

Cross sections and rate coefficients:

Arnaud & Raymond, 1992, Mazzotta et al., 1998; Kovaltsov et al. 2001

4

8

12

16

20

24

28

0.01 0.1 1 10

CONeMgSiFe

Energy (MeV/nuc)

Te = 1.2 10

6

Te = 2 106

Te = 1 107

Ostryakov et al., 1999; Barghouty & Mewaldt, 1999; Kocharov et al., 2000

Klecker et al., 2005


Comparison of Ionic Charge States with Stripping ModelII. The Non-Equilibrium Case

The Non-Equilibrium Case

Impact ionization by p + e

Radiative + dielectronic recombination

1) Qm depends on N*t

2) Equilibrium will be reached for

N * t ~ 1-10 * 1010 cm-3 s

(for E ~ 0.1 - 10 MeV/n)

3) Equilibrium N*t is energy

dependent

Kocharov et al., 2000

Q

24

20

16

12

8


Comparison of Fe Ionic Charge State Datawith Stripping Model

The Equilibrium Case

1. Qm at E < 0.1 MeV/n consistent with Te 1.2 - 1.4 106 K

2) Large Increase of Qm at E > 0.1 MeV/n

N * t ~ 1 * 1010 cm-3 s

t ~ 1 - 100 s: N ~ 108 - 1010 cm-3

-> Acceleration low in Corona

3) Increase of Qm with E larger than

in equilibrium stripping model

What is missing?

Klecker et al., 2005

8

10

12

14

16

18

20

22

24

0.01 0.1 1

Event 1Event 2Event 3Event 4STOF-AVG

Te=1.2 106 K

Te=1.4106 K

Averge Charge of Fe

Energy (MeV/nuc)


INTERPLANETARY TRANSPORTINCLUDING THE EFFECTS OF

Model, including acceleration Kartavykh et al., 2005

DIFFUSION CONVECTIONADIABATIC

DECELERATIONSOURCE

Energy Loss by Adiabatic Deceleration

1/E dE/dt = 4/3 Vsw / r s-1

Integrated (0.01 AU -> 1AU) energy loss depends on scattering mean free path and particle velocity.


A MODEL FOR ACCELERATION AND TRANSPORT

Acceleration Model, including

At the Sun: Spatial and Momentum Diffusion,Ionization, Coulomb Losses

Interplanetary Space: Transport, including Spatial Diffusion, Convection, Adiabatic Deceleration.

Simultaneous fit of: Energy Spectra Intensity-time

profile QFe (E)

(Kartavykh et al., 2004, 2005)


MODEL FITS FOR Ne, Mg, Si and Fe

July 3, 1999 Event July 20, 1999 Event


THE ENERGY DEPENDENCE OF THE IONIC CHARGE 2. Effect of Energy Spectra with M/Q-dependent Roll-Over

Klecker et al, 2001

10-8

10-6

10-4

10-2

100

102

104

10-2 10-1 100 101 102

61014186101418

Fe Flux (relative units)

Energy (MeV/nuc)

Eo (Fe10+) = 0.2 MeV/n

Eo (Fe10+) = 2.0 MeV/n

Fe Charge

0

5

10

15

20

10-1 100 101

Q (E0=0.2)

Q (E0=0.5)

Q (E0=1)

Mean Ionic Charge

Energy (MeV/nuc)

= 1.0

Assumed Energy Spectra J(E) ~ E−γ ex p (-E/E0)with E0 (A/Q) = E0 (proto )n * (Q/ )A

(Ellison & Ramaty, 1985, Tylka e t a., l 2000)

Fe Mean Ionic Charge computed with sample SW-Fe ionic charge distribution

Sinaia, September 6-10, 2005 31Mixing SW with QFe> 16+ from Impulsive EventsTylka et al. 2001

THE ENERGY DEPENDENCE OF THE IONIC CHARGE

3. Mixing of 2 Populations


SUMMARY-1Impulsive Events

• All non Interplanetary Shock related 3He-rich, Fe-rich events investigated so

far show

Qm (Fe) ~11 - 13 at 10 - 100 keV/n with a steep increase of Qm (Fe) to

Qm (Fe) ~14 - 20 in the energy range 180 - 550 keV/n.

• For several events, the increase above ~200 keV/n is steeper than expected for

charge stripping equilibrium conditions. Interplanetary transport effects

(adiabatic deceleration) are important and can explain the steeper increase.

• Homogeneous models provide good fits, if Q(E) is not too steep

• Inhomogeneous models are required to explain observations of steeper charge

spectra


SUMMARY-2

• The steep increase of Q with E for E < 1 MeV/nuc requires acceleration low

in the corona

N * A ~ 1-10 * 1010 cm-3 s

• For A ~ 10-100 s this corresponds to N ~ 108-1010 cm-3, i.e. altitudes < 2 Rs

High Charge States (e.g. Fe+20) observed at energies of ~ 1 MeV/n

can be used as Tracer for a Source Low in the Corona


SUMMARY-3Gradual Events

• High Charge States (and abundance enhancements) of Fe at Energies of

~ 1 MeV/nuc

Acceleration low in the corona

• High Charge States (and abundance enhancements) of Fe at Energies

> 10 MeV/nuc

Option 1: Injection and acceleration in the contemporary flare

Option 2: Injection and acceleration of 2 components by CME driven

coronal shock

(1) ~ solar composition, SW charge states

(2) ‘flare’ composition (heavy ion rich, high charge states)

Sinaia, September 6-10, 20051 Berndt Klecker Max-Planck-Institut für extraterrestrische Physik, 85741 Garching, Germany Workshop on Solar Terrestrial Interactions.

Documents