33 RD I NTERNATIONAL C OSMIC R AY C ONFERENCE , R IO DE JANEIRO 2013
T HE A STROPARTICLE P HYSICS C ONFERENCE
Inferred Ionic Charge States for Solar Energetic Particle Events with ACE and
STEREO
A. W. L ABRADOR1 , L. S. S OLLITT2 , C. M. S. C OHEN1 , A. C. C UMMINGS1 , R. A. L ESKE1 , G. M. M ASON3 , R. A.
M EWALDT1 , E. C. S TONE1 , T. T. VON ROSENVINGE4 , AND M. E. W IEDENBECK5
1
California Institute of Technology, Pasadena, CA 91125 USA
Department of Physics, The Citadel, Charleston, SC 29409 USA
3 JHU/Applied Physics Lab, Laurel, MD 20723 USA
4 NASA/Goddard Space Flight Center, Greenbelt, MD 20771 USA
5 Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109 USA
2
[email protected]
Abstract: Solar energetic particle (SEP) mean ionic charge states reflect conditions of the source material,
acceleration, and transport. These conditions can include temperatures at the source, the existence of highly
ionized seed material, or stripping during transport. In this paper, we examine the application of a method for
estimating mean ionic charge states for SEP events from energy-dependent decay timescales from the exponential
decays of SEP intensities. The method is applied to SEP event data from the Advanced Composition Explorer
(ACE) and the Solar Terrestrial Relations Observatory (STEREO), allowing multi-spacecraft observations of
individual SEP events.
Keywords: Solar Energetic Particles, Ionic Charge States.
1
Introduction
Solar energetic particle (SEP) mean ionic charge states can
reflect conditions of the source material as well as conditions during acceleration and transport. High mean ionic
charge states of SEPs may indicate, for example, elevated
plasma temperatures at the source, additional stripping during acceleration or transport, or acceleration of highly ionized seed material left over from prior hot sources or stripping conditions [1, 2, 3, 4]. Multi-spacecraft estimates of
charge states can potentially demonstrate longitudinal dependence in measured mean charge states arising from different source populations at different locations or from
magnetic connection effects during transport.
The SAMPEX (Solar Anomalous and Magnetospheric
Particle Explorer) spacecraft carried SEP detecting instruments that, in combination with the geomagnetic rigidity cutoff measurements during SAMPEX’s polar orbit,
could yield measurements of mean ionic charge states of
SEPs at energies up to ∼ 90 MeV/nuc for Fe (e.g. from
the Mass Spectrometer Telescope (MAST) aboard SAMPEX) [5]. This technique could yield mean ionic charge
state measurements at energies much higher than those obtained with electrostatic deflection measurements used by,
for example, the Solar Energetic Particle Ionic Charge Analyzer (SEPICA) aboard the Advanced Composition Explorer (ACE) [6]. (The geomagnetic rigidity cutoff technique was also used by the Low-Energy Ion Composition Analyzer (LICA) and the Heavy Ion Large Telescope
(HILT) aboard SAMPEX [7]).
With the inactivation of the SAMPEX/MAST instrument and the eventual atmospheric reentry of the SAMPEX spacecraft itself, such measurements of high energy
(up to ∼ 100 MeV/nuc) mean ionic charge states are not
currently available, but other indirect techniques for estimating mean ionic charge states remain.
Sollitt et al. [8, 9] demonstrated a method for inferring
mean ionic charge states from energy- and time-dependent
measurements of SEP intensities obtained by the Solar Isotope Spectrometer (SIS) aboard ACE. The SIS instrument
can detect Fe up to ∼ 140 MeV/nuc, for example. The analysis can also be supplemented at low energies by measurements from the Ultra Low Energy Isotope Spectrometer
(ULEIS) aboard ACE.
2 The Method
The method and the associated physical model are described in detail in Sollitt et al. [8, 9]. In brief summary, the
technique models an SEP event (e.g. a coronal mass ejection (CME)) as filling a magnetically bounded containment
volume in the heliosphere. After the SEP injection into the
containment volume is complete, SEPs diffuse out of the
containment volume, and this diffusive escape is seen as
exponential decays vs. time in SEP particle intensity. Additionally, the decay time constants are expected to be both
energy- and charge-dependent, fitting the form
1
1
=
+ W (αX E)γ
(1)
τ
τC
where τ is a decay time constant for some element X at
energy per nucleon E, τC is a low E decay time due to
convection and adiabatic cooling, γ is a constant that arises
from the rigidity dependence of the mean free path and W
is a normalization factor that applies to all particles species.
The parameter αX relates the mean ionic charge state of
element X to that of some calibration element (e.g. C, with
an assumed charge state of 5.9) and to the mean masses of
the two elements by the equation
2γ −1
QC AX γ
αX =
(2)
AC QX
where Q and A are mean ionic charge states and atomic
mass number, respectively (for C or for element X). Fi
Ionic Charge States for SEPs with ACE and STEREO
33 RD I NTERNATIONAL C OSMIC R AY C ONFERENCE , R IO DE JANEIRO 2013
Figure 1: Oxygen intensities vs. time for the 6 December
2006 SEP event as measured by ACE/SIS, for energy bins
ranging from 7.3-10 MeV/nuc (top) to 64-90 MeV/nuc
(bottom).
nally, the method described by Sollitt et al. [8, 9] involves
measuring exponential decay timescales for various elements at multiple energies and then fitting these decay
timescales vs. energy according to Equation 1, with free
parameters W, γ , and the αX ’s (one per element). Fits are
obtained with the downhill simplex of Nelder and Mead
[10] (the “amoeba”), and Equation 2 is then used to estimate mean ionic charge states for the measured elements.
This analysis is now extended to measurements from
the Low Energy Telescope (LET) aboard the Solar Terrestrial Relations Observatory (STEREO) [11]. The LET instrument has energy ranges that, while not as high as those
for SIS, overlap somewhat. For example, the LET stopping
energy ranges for C and Fe are ∼ 3.2 − 33 and ∼ 2.7 − 52
MeV/nuc, respectively, while those of SIS for C and Fe are
∼ 6.4 − 76 and ∼ 10.7 − 167 MeV/nuc.
3
Results
The 6 December 2006 SEP event is, significantly, the first
large SEP event during which, not only were the SAMPEX
and ACE spacecraft active, but also the STEREO spacecraft were operating near Earth. Measurements at this time
period could be important for establishing calibration between the spacecraft for these indirect ionic charge state
measurements. (Sollitt et al. [8, 9] already demonstrated
reasonable agreement between ACE indirect charge state
measurements and SAMPEX mean ionic charge state measurements for three SEP events from 1997 to 2001.)
Figure 1 shows an example of oxygen intensities vs.
time, as measured by ACE/SIS for the 6 December 2006
SEP event. For this analysis, the time period starting at
approximately 9 December 2006 (DOY’06 343) was used
to measure the exponential decay timescales. As can be
Figure 2: Measured decay timescales as a function of energy for C - Fe at various ACE/SIS energies, fitted to Equation 1. (C = black triangles, N = dark blue diamonds, O =
light blue diamonds, Ne = green diamonds, Mg = yellow
diamonds, Si = red diamonds, and Fe = black diamonds.)
The lower panel shows the difference between the measurements and the fit.
seen from the intensity data, the higher energy bins decay
faster, as expected for the model.
Figure 2 shows the results of fitting ACE/SIS data from
the 6 December 2006 SEP event to Equation 1, for a variety
of elements from C to Fe and for energies up to ∼ 90
MeV/nuc (for Fe). The lower panel of the figure shows
the difference between the measurements and the fit. Both
panels together demonstrate that this event was very well
fit by the analysis. The final fit parameters for this event are
γ = 1.23 ± 0.04, τC = 1.90 ± 0.16 (days), and W= (1.05 ±
0.13)x10−2 (units of 1/(days x (MeV/nuc)γ )).
Investigation of the 6 December 2006 event in
STEREO/LET data is still ongoing and incomplete as of
this writing, but we may at least compare the ACE/SIS result with that from SAMPEX/MAST for this event. The
ACE/SIS indirect mean ionic charge state measurements
for elements from N to Fe are shown in Figure 3, with
the SAMPEX/MAST results for this event also shown for
comparison [12]. Both results are statistically consistent
for most elements.
We have attempted fits to several other SEP events with
STEREO/LET data alone, and one preliminary result is
Ionic Charge States for SEPs with ACE and STEREO
33 RD I NTERNATIONAL C OSMIC R AY C ONFERENCE , R IO DE JANEIRO 2013
Figure 3: Mean ionic charge states as a function of nuclear
charge for N, O, Ne, Mg, Si, and Fe, during the 6 December 2006 SEP event. The SAMPEX/MAST results are
shown as open circles, and the ACE/SIS indirect measurements are shown as black circles, slightly offset.
shown in Figure 4, for the 27 January 2012 event with
the STEREO Ahead spacecraft. The low abundance of Fe
in this event detected by LET Ahead in the time period
analyzed made it difficult to obtain a significant measure
of the mean Fe charge state. However, the analysis itself
did successfully yield charge state estimates for the lighter
elements.
4
Discussion and Conclusion
We have done initial work in applying this method to a sample of SEP events from 2006 to 2012, in both ACE/SIS and
STEREO/LET data, with additional events from 2013 to
be attempted by the conference. The sample of SEP events
has not been exhaustive, but we may already conclude that
the fraction of SEP events that yield charge state estimates
by this technique is small. Sollitt et al. [8, 9] noted that, of
40 events studied from the early years of the ACE mission,
only 4 yielded charge states by this method, and our preliminary attempts result in similarly low yields.
A primary difficulty appears to be in finding SEP events
with useful counting statistics during the exponential decay phase of the events. In SEP events in which the exponential decay phase occurs well after the peak in intensity,
the passage of shock fronts can distort the exponential decay phase of the event or delay it until sufficiently late that
counting statistics are too low to be useful. Additionally,
the increasing distance between ACE and the STEREO
spacecraft makes events large enough to yield charge estimates at both ACE and one of the STEREO spacecraft
unlikely to occur. However, as the STEREO spacecraft approach each other (already at ∼ 86◦ separation), there will
be opportunities to compare measurements between them.
There are additional events from 2006-2013 that remain
Figure 4: Preliminary mean ionic charge state estimates
for the 27 January 2012 SEP event, using STEREO/LET
Ahead data. Uncertainties are smaller than plot symbols
for some elements.
to be investigated with this method, and these may be reported at the conference. Future work will also incorporate
lower energy ACE/ULEIS data and STEREO/SIT data.
Acknowledgment: This work has been supported by NASA
Grants NNX13AH66G and NAS5-03131.
References
[1] Mason, G.M. et al. The Astrophysical Journal Letters 748
(2012) L31 doi:10.1088/2041-8205.
[2] Moebius, E., et al. 2003, in Proc. 28th Int. Cosmic Ray
Conf. (Tsukuba), 6, 3273.
[3] Klecker, B., et al., Space Science Reviews, 130 (2007) 273
doi:10.1007/s11214-007-9207-1.
[4] Mewaldt, R.A., et al., Space Science Reviews, 171 (2012)
97 doi:10.1007/s11214-012-9884-2.
[5] Leske, R.A. et al., The Astrophysical Journal Letters 452
(1995) L149 doi:10.1086/309718.
[6] Moebius, E., et al., Space Science Reviews, 86 (1998) 447.
[7] Oetliker, M. et al., The Astrophysical Journal, 477 (1997)
495 doi:10.1086/303679.
[8] Sollitt, PhD Thesis, California Institute of Technology
(2004), http://thesis.library.caltech.edu/3493/.
[9] Sollitt, L.S. et al., The Astrophysical Journal 679 (2008)
910-919 doi:10.1086/587121.
[10] Nelder, J. A. and Mead, R.. ”A simplex method for
function minimization”. Computer Journal 7 (1965) 308313.
doi:10.1093/comjnl/7.4.308.
[11] Mewaldt, R.A. et al., Space Science Reviews, 136 (2008)
285-362 doi:10.1007/s11214-007-9288-x.
[12] Labrador, A.W. et al. (2012), New Estimates of Inferred
Ionic Charge States for Solar Energetic Particle Events with
ACE and STEREO, Abstract SH21A-2165 presented at
2012 Fall Meeting, AGU, San Francisco, Calif., 3-7 Dec..