Solar flare observations with INTEGRALSPI - PowerPoint PPT Presentation

1 / 26
About This Presentation
Title:

Solar flare observations with INTEGRALSPI

Description:

Photosphere. No PAS. Strong PAS. The magnetic loop model. IEEC, Barcelona, September 23, 2004 ... PHOTOSPHERE. CORONA. isotropic accelerated- particle release ... – PowerPoint PPT presentation

Number of Views:36
Avg rating:3.0/5.0
Slides: 27
Provided by: namep52
Category:

less

Transcript and Presenter's Notes

Title: Solar flare observations with INTEGRALSPI


1
Solar flare observations with INTEGRAL/SPI
(M. Gros, J. Kiener, V. Tatischeff et al.)
Nuclear g-rays
e-
p, a...
Hard X-rays
Neutrons
CORONA
TRACE RHESSI, 28-Oct-2003
2.22 MeV g-ray line
CHROMOSPHERE
n
PHOTOSPHERE
H
12C
12C
p
  • Nuclear de-excitation lines

hn
IEEC, Barcelona, September 23, 2004
2
The INTEGRAL satellite
  • Scientific objectives AGN, g-ray bursts,
    compact objects, novae, SNe, interstellar g-ray
    emissions...
  • Launched (Proton) on 17 Oct 2002

IEEC, Barcelona, September 23, 2004
3
Interest of SPI for solar flare physics
  • Compact array of 19 hexagonal Ge detectors
    (Stot500 cm2) good efficiency at high energy
    (compared to RHESSI) using "multiple events"
  • Anti-Coincidence veto System (ACS) of 91 BGO
    scintillator crystals Spro60009000 cm2

IEEC, Barcelona, September 23, 2004
4
SPI observations of the 2003 Oct 28 solar flare
(X17.2)
  • During INTEGRAL observation of IC443 (rev 127
    PI A. Bykov)
  • Simulated response function for the satellite
    configuration during the flare in progress
    (Weidenspointner et al.)
  • All results are preliminary

IEEC, Barcelona, September 23, 2004
5
Measured spectra and time history
pair prod.
mostly instrumental
  • With all types of Ge events (including multiples
    2-5)

IEEC, Barcelona, September 23, 2004
6
4.44 and 6.13 MeV line characteristics
0.95 - 0.65
0.23 - 0.22
0.24 - 0.29
1.11 - 0.83
  • RHESSI results are for the 23 July, 2002 X4.8
    flare (73 helio. angle) - Smith et al. 2003

IEEC, Barcelona, September 23, 2004
7
4.44 and 6.13 MeV line shape calculations
  • Detailed model based on laboratory data
  • Sensitive to the angular distribution of the
    accelerated particles and the a/p ratio
  • Best fit results
  • 12C only a/p 0.00 DQ 20
  • 16O only a/p 0.09 DQ 34
  • 12C 16O a/p 0.03 DQ 29

IEEC, Barcelona, September 23, 2004
8
The 6.92 and 7.12 MeV lines of 16O
  • Fit with a fixed line shape same relative
    redshift and FWHM as for the 6.13 MeV line
  • The two 16O lines at 7 MeV are resolved for
    the first time
  • From a simplified model of solar g-ray
    absorption

IEEC, Barcelona, September 23, 2004
9
Gamma-ray line ratios
  • Fast ion composition Solar Energetic Particles
    (SEP) from impulsive flares
  • Fast ion energy spectrum dN/dE ? E-S
  • Nuclear de-excitation lines (thick target
    production model) compared to 2.22 MeV line
    production (Hua et al. 2002) ? DS

?
Smin for a/p0.1
Smax for a/p0.1
IEEC, Barcelona, September 23, 2004
10
With a stochastic acceleration spectrum
The source spectrum should be a modified Bessel
function rather than a power law (e.g.
Forman et al. 1986). ? no improvement for C/O a
acceleration efficiency T escape time from
the acceleration region
?
IEEC, Barcelona, September 23, 2004
11
With SMM and OSSE data
  • 9 SMM flares with strong (and complete) g-ray
    line emission (SM95)
  • OSSE 1991 June 4 flare (Murphy et al. 1997)
  • RHESSI results not yet taken into account

DS determination for the 1989 Nov 15 flare
Correction for heliocentric angle
?
IEEC, Barcelona, September 23, 2004
12
The 12C/16O line ratio problem
from Ramaty et al.
  • Calculated F(4.44)/F(6.13) overestimates by a
    factor of 1.5 the average line ratio obtained
    from SMM, OSSE and SPI data.
  • Origin of the problem - the
    interaction model ? - the cross
    sections ? - the abundances of
    12C and 16O in the ambient medium (coronal,
    from gradual event SEP) ?
  • comparison with the 2 other significant lines
    detected with SMM and OSSE at 1.37 (24Mg) and
    1.63 MeV (20Ne)

IEEC, Barcelona, September 23, 2004
13
Cross sections (1)
Mainly from KMR02 (ApJ Suppl), the figures.
  • 4.44 MeV line S3 S4.5
  • a 12C(p,p)12C 47.4 43.1
  • b 14N(p,x)12C (?) 1.6 0.2
  • c 16O(p,x)12C 35.7 9.5
  • d 12C(?,?)12C 8.7 39.3
  • e 14N(?,x)12C (?) 0.4 0.3
  • f 16O(?,x)12C 6.3 7.5
  • A(b,c)D cross section measured by the g-ray
    method (1020 uncertainties)

(with a/p0.1)
(?) Cross sections overestimated in KMR02
calculated with EMPIRE-II (nuclear statistical
model)
IEEC, Barcelona, September 23, 2004
14
Cross sections (2)
  • 6.13 MeV line ? 6.129 MeV (16O) 6.175 MeV
    (15O),
  • but not the 6.322 MeV line (15N), see
    Mandzhavidze et al. (1999).
  • S3 S4.5
  • a 16O(p,p)16O 67.0 42.7
  • b 20Ne(p,x)16O 6.0 1.6
  • c 16O(p,x)15O 11.0 0.5
  • d 16O(?,?)16O 15.2 54.4
  • e 20Ne(?,x)16O (?) 0.6 0.8
  • f 16O(?,x)15O lt0.1 lt0.1

(?) Cross section not considered in KMR02,
calculated with EMPIRE-II
IEEC, Barcelona, September 23, 2004
15
Cross sections (3)
  • 7 MeV lines ? 6.92 MeV 7.12 MeV (16O)
  • S3 S4.5
  • a 16O(p,p)16O6.92 78.0 40.0
  • c 16O(?,?)16O6.92 22.0 60.0
  • b 16O(p,p)16O7.12 87.1 52.3
  • d 16O(?,?)16O7.12 12.9 47.7
  • Minor contributions (neglected) from
    20Ne spallation (EMPIRE-II)

IEEC, Barcelona, September 23, 2004
16
Cross sections (4)
  • 1.63 MeV line ? 1.634 MeV (20Ne) 1.636 MeV
    (23Na) 1.635 MeV (14N)
  • S3 S4.5
  • a 20Ne(p,p)20Ne 55.1 62.1
  • b 24Mg(p,x)20Ne,23Na 20.5 4.1
  • c 28Si(p,x)20Ne 5.3 0.5
  • d 20Ne(?,?)20Ne 7.0 27.1
  • e 24Mg(a,x)20Ne,23Na 2.5 1.3
  • a 14N(p,p)14N 4.1 2.9
  • b 16O(p,x)14N 4.9 0.3
  • c 14N(?,?)14N 0.6 1.8

IEEC, Barcelona, September 23, 2004
17
Cross sections (5)
  • 1.37 MeV line ? 1.369 MeV (24Mg) 1.370 MeV
    (55Fe) 1.367 MeV (59Ni)
  • S3 S4.5
  • a 24,25,26Mg(p,x)24Mg 85.8 74.5
  • b 28Si(p,x)24Mg 7.2 0.9
  • c 56Fe(p,x)55Fe 1.1 0.1
  • d 24Mg(?,?)24Mg 5.3 22.2
  • e 56Fe(a,n)59Ni 0.6 2.3

IEEC, Barcelona, September 23, 2004
18
With the 1.63 and 1.37 MeV lines
  • etheory20 (due to s) added in quadrature to
    edata for the ?2 probabilities
  • Goodness-of-fits
  • Ambient medium ? coronal
  • but ASEP(C) is too high
  • a/p0.1 is favored. Then Ne/O?0.15 and Mg/O?0.20

The Dec 16, 1988 Flare. Not included in the
probability calculations. Ambient ? photosph. ?
IEEC, Barcelona, September 23, 2004
19
With DS from F2.22/F6.13 only
  • same results, but on average the probabilities
    are slightly lower as DS?

IEEC, Barcelona, September 23, 2004
20
The C abondance in the interaction region
With a/p0.1
(C/O)SEP0.46?0.01 (Reames 1999) (C/O)pho0.50?0.
08 (Lodders 2003)
  • Good consistency of the 3 probability
    distributions
  • From maximum likelywood (C/O) 0.28 ? 0.03
    (1s) 0.28 ? 0.08 (2s)

IEEC, Barcelona, September 23, 2004
21
A new photospheric C abundance ?
speculative
  • (C/O)chr0.3 but (C/O)pho0.5 ?
  • Apho(C) and Apho(O) are uncertain recent
    substantial revisions (NLTE, 3D models)
  • A reduced Asol(C) would better fit the C
    abondance gradient in the Galactic disk (see Hou
    et al. 2000, fig. 6)
  • Anders Grevesse (1989)
  • Grevesse Sauval (1998)
  • Holvecker (2001)
  • Lodders (2003)
  • Asplund et al. (2004), AA for O, in prep. for C

for a/p0.1
IEEC, Barcelona, September 23, 2004
22
The photospheric 3He abundance
Neutrons
2.22 MeV
n
ne
n
n
3He
p
e-
H
p
3H
  • The time evolution of the 2.22 MeV line emission
    is sensitive to Apho(3He)
  • s3He(n,p)3H?1.6104?s1H(n,g)2H
  • ? tNRC 1 / n(3He)sNRCvn
  • tRC ? (H/3He) ? 6.2510-5
  • Neutron-production time history ? prompt g-ray
    line emission (good quality data with SPI)
  • Not measured by atomic spectroscopy

IEEC, Barcelona, September 23, 2004
23
The magnetic loop model
(Hua, Lingenfelter, Murphy, Ramaty...)
isotropic accelerated- particle release
constant B
  • MHD turbulence ? pitch-angle scattering

CORONA
magnetic mirroring (sin2? ? B)
CHROMOSPHERE
PHOTOSPHERE
B ? (pressure)d
loss cone"
  • No PAS (mean free path ? ? ?) fan beam of
    interacting particles (i.e. parallel to the
    solar surface)
  • Strong PAS loss cone continuously repopulated ?
    downward beam

Hua et al. (2002)
IEEC, Barcelona, September 23, 2004
24
Calculated 2.22 MeV lightcurves
  • Monte-Carlo code (Hua et al. 1987, 2002) to
    simulate (i) the propagation and interaction
    of the accelerated particles (ii) the neutron
    production and propagation (iii) the 2.22 MeV
    line production and absorption
  • For instantaneous release of the accelerated
    particles, the 2.22 MeV lightcurves fall faster
    with increasing PAS (decreasing l) and increasing
    3He/H (see Murphy et al. 2003)

IEEC, Barcelona, September 23, 2004
25
The photospheric 3He abundance results
? fan beam
downward beam ?
  • The two free parameters are strongly correlated
  • l from 4.44 and 6.13 MeV line shapes ? more
    accurate 3He/H
  • Solar neutron measurements (monitors
    CORONAS/SONG) could help...

IEEC, Barcelona, September 23, 2004
26
Summary
  • From g-ray spectroscopy of the 2003 Oct 28 solar
    flare with SPI
  • - energy spectrum of the accelerated ions (g-ray
    line fluences)
  • - accelerated a/p ratio (g-ray line shapes and
    fluences)
  • - amount of PAS in magnetic loop/angular
    distribution of the interacting particles (g-ray
    line shapes and 2.22 MeV lightcurve)
  • ? acceleration and transport processes
  • - ambient C abundance (g-ray line fluences)
  • - ambient 3He abundance (2.22 MeV lightcurve)
  • ? solar composition and atmospheric response
  • Much more to do
  • - timing analyses using the ACS (and radio data)
  • - analyses of the 2003 Nov 4 flare (near the
    solar limb !)
  • ...

IEEC, Barcelona, September 23, 2004
Write a Comment
User Comments (0)
About PowerShow.com