1

Introduction to the SHINE Campaign EventsIntroduction to the SHINE Campaign Events2002 April 21 and 2002 August 242002 April 21 and 2002 August 24

Allan J. TylkaAllan J. Tylka

Code 7652, Naval Research Laboratory, Washington DC 20375Code 7652, Naval Research Laboratory, Washington DC 20375

allan.tylka@nrl.navy.mil; 202-767-2200allan.tylka@nrl.navy.mil; 202-767-2200

SHINE 2004SHINE 2004

Plenary Session on the SHINE Campaign EventsPlenary Session on the SHINE Campaign Events

Big Sky, MontanaBig Sky, Montana

2004 June 282004 June 28

2

OutlineOutline

1.1. Objective: What do we hope to learn by comparing these two events?Objective: What do we hope to learn by comparing these two events?

2.2. Review: Characteristics of these two events Review: Characteristics of these two events

3.3. Campaign StrategyCampaign Strategy

Solar observationsSolar observations

SEP StudiesSEP Studies

4.4. Hypotheses about the SEP PhysicsHypotheses about the SEP Physics

5.5. Additional Facts:Additional Facts:

Based on statistical study of the 44 largest SEP events of Cycle Based on statistical study of the 44 largest SEP events of Cycle 2323

-- selected on the basis of >30 MeV proton fluence-- selected on the basis of >30 MeV proton fluence

-- no heavy-ion biases-- no heavy-ion biases

Provide context for our deliberations.Provide context for our deliberations.

6.6. Recent Modeling EffortsRecent Modeling Efforts

7. Posters7. Posters

3SEP Variability at High EnergiesSEP Variability at High Energies(above a few tens of MeV/nuc)(above a few tens of MeV/nuc)

Our Our Objective:Objective:

Discover Discover what makes what makes these two these two events events different at different at high high energies!energies!

Fe/C (normalized to corona) vs. EnergyFe/C (normalized to corona) vs. Energy

(only GLE of 2002)(only GLE of 2002)

(biggest event of 2002)(biggest event of 2002)

Data from ACE/ULEIS, ACE/EPAM, Wind/LEMT, & ACE/SISData from ACE/ULEIS, ACE/EPAM, Wind/LEMT, & ACE/SIS

4Whatever makes these events different, Whatever makes these events different, it probably happens very near the Sun…it probably happens very near the Sun…

2001 April 15 Ground Level Event2001 April 15 Ground Level Event(like 2002 August 24, but bigger)(like 2002 August 24, but bigger)

SEP onsets & peak of ~GeV production SEP onsets & peak of ~GeV production

occurred when the CME was at ~2 Roccurred when the CME was at ~2 RSS..

5Why these two events?Why these two events?

Similar solar sources -- a ‘controlled’ experiment!Similar solar sources -- a ‘controlled’ experiment!

21 Apr 200221 Apr 2002 24 Aug 200224 Aug 2002

CME (SOHO, S. Yashiro):CME (SOHO, S. Yashiro):

Speed (km/s)Speed (km/s) 2409 km/s2409 km/s 1878 km/s1878 km/s

Flare (NOAA/SGD):Flare (NOAA/SGD):

SizeSize X1.5 / 1FX1.5 / 1F X3.1 / 1FX3.1 / 1F

LocationLocation S14S14oo W84 W84oo S02S02oo W81 W81oo

SW Speed* at 1 AU (Wind, K. Ogilvie)SW Speed* at 1 AU (Wind, K. Ogilvie) 452-504 km/s452-504 km/s 345-436 km/s345-436 km/s

Nominal Magnetic Connection LongitudeNominal Magnetic Connection Longitude W46W46oo-W51-W51oo W52W52oo-W66-W66oo

Associated Shock at 1 AU (ACE, C.W. Smith):Associated Shock at 1 AU (ACE, C.W. Smith):

Transit Time to 1 AU Transit Time to 1 AU 51 hours51 hours 58 hours58 hours

Velocity Jump (km/s)Velocity Jump (km/s) ~200 km/s~200 km/s ~100 km/s~100 km/s

ESP increase?ESP increase? < 1 MeV only< 1 MeV only < 0.5 MeV only< 0.5 MeV only

* * During the first 12 hours.During the first 12 hours.

6Compare Spectra: Compare Spectra: 2002 April 212002 April 21 & & 2002 August 242002 August 24(Data from ACE/ULEIS, ACE/EPAM, Wind/LEMT, and ACE/SIS)(Data from ACE/ULEIS, ACE/EPAM, Wind/LEMT, and ACE/SIS)

2002 August 24:2002 August 24:

•Double power-laws:Double power-laws:

Fits to Band et al. (1993) fcnFits to Band et al. (1993) fcn

•At high energies:At high energies:

Fe harder than O (and C)Fe harder than O (and C)

2002 April 21:2002 April 21:

•Fits to Fits to F(E) ~ EF(E) ~ E-- exp(-E/E exp(-E/E00))

•At high energies:At high energies:

Fe softer than O (and C)Fe softer than O (and C)

Galactic Cosmic Galactic Cosmic RaysRays

7

Pre-Event Particle BackgroundPre-Event Particle Background

17-23 April 200217-23 April 2002 20-26 August 200220-26 August 2002

Nominal Fe/O ~ 0.1 in the Nominal Fe/O ~ 0.1 in the 4 days preceding the event4 days preceding the event

Enhanced Fe/O ~ 1 in the Enhanced Fe/O ~ 1 in the 4 days preceding the event4 days preceding the event

Blue: OxygenBlue: Oxygen Red: IronRed: Iron

These two CMEs likely launched into very different seed populations.These two CMEs likely launched into very different seed populations.

Data Data gapgap

8GOES ProtonsGOES Protons>10 MeV >10 MeV >50 MeV>50 MeV >100 MeV>100 MeV

21 April 200221 April 2002(Fe-poor at high energies)(Fe-poor at high energies)

•>10 MeV profiles very similar, except for normalization. >10 MeV profiles very similar, except for normalization. •But >50, >100 MeV profiles have different shapes in the two events.But >50, >100 MeV profiles have different shapes in the two events.•No increase in high-energy particles at shock arrival at Earth (arrows).No increase in high-energy particles at shock arrival at Earth (arrows).

24 August 200224 August 2002(Fe-rich at high energies)(Fe-rich at high energies)

9

Campaign StrategyCampaign Strategy

1.1. Solar ObservationsSolar Observations

• Identify differences in SOHO, TRACE, RHESSI, etc. observationsIdentify differences in SOHO, TRACE, RHESSI, etc. observations

• Brainstorm about their possible relevance to SEPsBrainstorm about their possible relevance to SEPs

-- suggest new hypotheses-- suggest new hypotheses

-- confirm or refute hypotheses motivated by SEP observations alone-- confirm or refute hypotheses motivated by SEP observations alone

• Two talks in this session:Two talks in this session:

David Alexander: Pre-Event History of these Active RegionsDavid Alexander: Pre-Event History of these Active Regions

John Raymond: UVCS Observations John Raymond: UVCS Observations

2.2. SEP StudiesSEP Studies

• WG3 Session Tuesday Morning:WG3 Session Tuesday Morning:

“ “What causes energy-dependent Fe/O in large SEP events?”What causes energy-dependent Fe/O in large SEP events?”

Invited talks by Desai, Giacalone, Kota, Li, and TylkaInvited talks by Desai, Giacalone, Kota, Li, and Tylka

10

Hypotheses about These EventsHypotheses about These EventsThe SEP differences between these two events are due to:The SEP differences between these two events are due to:

1.1. Shock Geometry and Seed PopulationShock Geometry and Seed Population

• August event involves a quasi-perp shock operating on a seed August event involves a quasi-perp shock operating on a seed population containing both flare & solar-wind ions.population containing both flare & solar-wind ions.

• April event comes from a seed population dominated by the solar April event comes from a seed population dominated by the solar wind & accelerated by a shock, most likely quasi-parallel.wind & accelerated by a shock, most likely quasi-parallel.

2.2. Seed Population OnlySeed Population Only

• August event had a strong flare component in its seed particles; the August event had a strong flare component in its seed particles; the April event did not.April event did not.

• Shock geometry effects are non-existent/irrelevant.Shock geometry effects are non-existent/irrelevant.

3.3. A Direct Flare Component at High EnergiesA Direct Flare Component at High Energies

• Both shock geometry and seed population are irrelevant.Both shock geometry and seed population are irrelevant.

• August event had a component August event had a component directlydirectly accelerated to high energies accelerated to high energies by the flare. by the flare.

• The flare component was missing or overwhelmed in the April event.The flare component was missing or overwhelmed in the April event.

4.4. Other ideas??Other ideas??

11

Additional “Facts”Additional “Facts”

• based on the examination of large samples of eventsbased on the examination of large samples of events

• provide additional constraints & challenges for our hypotheses.provide additional constraints & challenges for our hypotheses.

1.1. SEPs vs. ESPs (traveling shocks at 1 AU)SEPs vs. ESPs (traveling shocks at 1 AU)

2.2. Longitude Distribution of Solar Source RegionsLongitude Distribution of Solar Source Regions

3.3. Spectral CharacteristicsSpectral Characteristics

4.4. Average Fe/O Enhancement at High EnergiesAverage Fe/O Enhancement at High Energies

12SEPs and ESPs SEPs and ESPs Strong Energy Dependence in Fe/C (and Fe/O)Strong Energy Dependence in Fe/C (and Fe/O)

Starting point: 44 SEP events 1997-2004 and 55 ESP events 1997-2002Starting point: 44 SEP events 1997-2004 and 55 ESP events 1997-2002

In both cases, about 1/3 of the event sample showed strong energy dependence. In both cases, about 1/3 of the event sample showed strong energy dependence.

Same range of behavior in both SEP and ESP events (but at different energies).Same range of behavior in both SEP and ESP events (but at different energies).

Energetic Storm ParticlesEnergetic Storm Particles(accelerated locally, near 1 AU)(accelerated locally, near 1 AU)

From Desai et al. 2003 & Tylka et al. 2004

Solar Energetic ParticlesSolar Energetic Particles(accelerated near the Sun) (accelerated near the Sun)

From Tylka et al. 2004

13

Longitude Distribution of Longitude Distribution of SEP EventsSEP Events

•These large Fe-rich events are These large Fe-rich events are observed from sources all observed from sources all across the Sunacross the Sun

•Unlike classic Unlike classic 33He-rich He-rich “impulsive” events:“impulsive” events:

•observable only at observable only at magnetically-connected solar magnetically-connected solar longitudes:longitudes:

•85% between W30 and W80.85% between W30 and W80.

•Spread primarily due to Spread primarily due to variation in solar wind speed.variation in solar wind speed.

From Tylka et al. 2004

From Reames 1999

14Fe/O vs. Spectral SteepeningFe/O vs. Spectral SteepeningEvent-Integrated Oxygen SpectrumEvent-Integrated Oxygen Spectrum

Events with enhanced high-energy Fe/O tend to have spectra more like power Events with enhanced high-energy Fe/O tend to have spectra more like power laws.laws.

Events with suppressed high-energy Fe/O tend to have spectra that steepen at high Events with suppressed high-energy Fe/O tend to have spectra that steepen at high energies.energies.

Fe/O at 30-40 MeV/nuc v. Fe/O at 30-40 MeV/nuc v. 22 - - 11

Since the event selection was based on proton fluence, there are no observational biases in this plot.

15Average Fe/O Enhancement at High EnergiesAverage Fe/O Enhancement at High Energies

For events with enhanced For events with enhanced Fe/O at high energies, the Fe/O at high energies, the averageaverage Fe/O value “plateaus” Fe/O value “plateaus” at ~5x coronal.at ~5x coronal.

Significantly below the flare Significantly below the flare average of ~8x coronal.average of ~8x coronal.

Numbers on the datapoints tell how many events were used in the average.

30-40 MeV/nuc30-40 MeV/nuc 40-50 MeV/nuc40-50 MeV/nuc 50-75 MeV/nuc50-75 MeV/nuc

16

Summary of Additional FactsSummary of Additional Facts

1.1. The same types of extreme energy dependence in Fe/O are seen in both The same types of extreme energy dependence in Fe/O are seen in both high-energy SEPs (accelerated near the Sun) and ESP events (accelerated high-energy SEPs (accelerated near the Sun) and ESP events (accelerated by shocks near 1 AU, but at lower energies.)by shocks near 1 AU, but at lower energies.)

Suggests (but does not prove) a common origin in shock physics.Suggests (but does not prove) a common origin in shock physics.

2.2. SEP events with enhanced high-energy Fe/O are seen from a broad range SEP events with enhanced high-energy Fe/O are seen from a broad range of source longitudes.of source longitudes.

Not like “classic” flare-associated impulsive events, with Not like “classic” flare-associated impulsive events, with 33He/He/44He >10% He >10% at ~1 MeV/nucat ~1 MeV/nuc

3.3. Enhanced Fe/O is anti-correlated with spectral steepening at high energies.Enhanced Fe/O is anti-correlated with spectral steepening at high energies.

There are spectral as well as compositional differences to be explained.There are spectral as well as compositional differences to be explained.

4.4. For events with enhanced Fe/O at high energies, the For events with enhanced Fe/O at high energies, the averageaverage Fe/O value Fe/O value “plateaus” at ~5x coronal.“plateaus” at ~5x coronal.

This average is significantly below the impulsive average of ~8x This average is significantly below the impulsive average of ~8x coronal.coronal.

17The “Zoo” of Fe/O vs. EnergyThe “Zoo” of Fe/O vs. EnergyDataDataModeling Modeling (Lee & Tylka)(Lee & Tylka)

Modeling based on:Modeling based on:

Variable shock geometry (quasi-perp & quasi-parallel)Variable shock geometry (quasi-perp & quasi-parallel)

Variable seed population (SW & flare suprathermals)Variable seed population (SW & flare suprathermals)

R = 0

R = 0

R = 0.01

R = 0.1

R = 1.0

R = 10

R = 0

18

Posters on the 2002 Campaign EventsPosters on the 2002 Campaign Events

8. Alexander, David - SHINE SEP Campaign Events: Detailed Comparison of active regions AR9906 and AR0069 in the build-up to the SEP events of 21 Apr 2002 and 24 Aug 2002

9. Daou, Antoun G. - Transient activity in regions AR9906 and AR0069 in the build-up to the SEP events of 21 Apr 2002 and 24 AUG 2002

10. Liu, Rui - SHINE SEP Campaign Events: Global and Local Magnetic Field Evolution in Build-up to the SEP events of 21 Apr 2002 and 24 Aug 2002

11. Coyner, Aaron - SHINE SEP Campaign Events: Long-term development of solar corona in build-up to the SEP events of 21 Apr 2002 and 24 Aug 2002

12. Gopalswamy, Nat - Coronal and Interplanetary Shocks of 2002 April 21 and their Implication to the CME Onset Time

13. Cane, Hilary - The role of CMEs and their shocks in determining SEP abundances

14. Tylka, Allan et al. - Shock Geometry, Seed Populations, and the Origin of Variable Elemental Composition at High Energies in Large Gradual Solar Particle Events

15. Keller, Kristi - Validation of Heliosphere Models

16. Evenson, Paul - Spaceship Earth Observations of the 24 August 2002 Event

19SPACESHIP EARTHSPACESHIP EARTH OBSERVATIONS OF THE OBSERVATIONS OF THE24 AUGUST 2002 EVENT (a SHINE campaign event)24 AUGUST 2002 EVENT (a SHINE campaign event)

J. W. Bieber, P. Evenson, R. Pyle, D. Ruffolo, and T. KhumlumlertJ. W. Bieber, P. Evenson, R. Pyle, D. Ruffolo, and T. Khumlumlert

We analyze data collected by We analyze data collected by the 11-station the 11-station Space-ship Space-ship Earth Earth neutron monitor neutron monitor network to derive and model network to derive and model the evolution of the particle the evolution of the particle density and anisotropy in this density and anisotropy in this event.event.

Timing Results (at Sun): Start of Production of GeV-protons: 01:08 + 2 min

Estimated CME Location: 3.8 + 0.4 RS

For comparison: H: Start 00:47; Peak 00:55; End 01:15

20

FLARE (Fe/O>1) AND CME-SHOCK (Fe/O<1)

Eve

nt-

aver

aged

Fe/

O (

rela

tive

to

co

ron

a) Contributions of the two components vary with energy from event to event and observing location. Plot shows >10 MeV/nuc values for 45 events.

No in situ shock at 1 AU

Shock – no particle effect at 1 AU

Particle effect at shock passage

April 21 2002: strong shock component

August 24 2002: impulsive flare component dominant

POSTER #13 Hilary CANE

SEP COMPOSITION INDICATES TWO COMPONENTS

H