RHESSI Xray Spectral Analysis

1
RHESSI X-ray Spectral Analysis
Leeann Chau Dang1, Brian R. Dennis1, Richard A.
Schwartz1,2, and A. Kimberley Tolbert1,3
1 Goddard Space Flight Center, Greenbelt, MD
20771 Brian.R.Dennis_at_nasa.gov2 The Catholic
University of America, Washington, DC 20064
richard.a.schwartz.1_at_gsfc.nasa.gov3 RSIS,
Lanham, MD Anne.K.Tolbert.1_at_gsfc.nasa.gov
Abstract RHESSI X-ray spectroscopy is possible
down to the lowest energies (3 to 6 keV
depending on attenuator state) provided that the
many instrumental effects are adequately
accounted for. Fully exploiting these
observations, especially at the lower end of the
RHESSI spectral range below 20 keV, has proven
to be challenging given the complexity of the
instrument response and the many effects that
must be considered in determining the incident
photon spectrum from the measured count-rate
spectrum. These effects include the strong
attenuation of the material (including the
attenuators when in place) overlying the
detectors, the instrument background lines from
the germanium detectors and the tungsten grids,
the K-escape process and other mechanisms that
result in lower than expected pulse amplitudes,
electronic effects at high count rates such as
pulse pile-up and change in gain,
detector-to-detector differences, variations in
the non-flare background spectrum, etc. We will
discuss the current status of our ability to
estimate the magnitude of all of these effects
and to determine the properties of the incident
photon spectrum using the forward-fitting method.
The introduction of Chianti version 5.2 has
allowed us to fit expected thermal continuum and
line spectra with a single temperature or a given
distribution of temperatures and to determine the
iron abundance from the intensity of the
iron-line complex peaking at 6.7 keV. Pulse
pile-up and albedo corrections can now be applied
during the spectral fitting process allowing us
to determine best-fit estimates of both the
thermal and nonthermal components of the photon
spectrum coming directly from the X-ray
source(s). Temperatures (or temperature
distributions) down to as low as 8? MK can be
reliably estimated. While the separation of the
thermal and nonthermal components remains
difficult, utilizing the new ability to reliably
measure the photon spectrum below 10 keV combined
with the temporal and spatial signatures make it
possible to resolve this issue in specific cases
with greater reliability than has previously been
possible.
RHESSI
SOXS
(Solar X-ray Spectrometer) on Indian Telescope
Detector Parameters
Detector type Hyperpure Germanium Resolution
1 keV (FWHM) Threshold 3 keV Det. 2 bad
resolution, 20 keV threshold pre-2005? Det. 5
poor resolution Det. 7 poor resolution, 10 keV
threshold
Detector type Silicon (PIN) Resolution 0.8
keV (FWHM) Energy range 4 to 25 keV
RHESSI SOXS GOES Comparison
Plot showing comparison of the emission measure
values at T 2 keV obtained from the
multithermal spectral fits to the RHESSI and SOXS
data. The GOES emission measures are shown vs.
time for single temperature fits.
Background Line at 8.3 keV
Fig. 1
Fig. 2
Comparison between RHESSI and SOXS count-rate
spectra. Figure 1 is from the peak of the flare
and figure 2 is from the decay. The black
histograms represent the SOXS background-subtracte
d count-rate spectra. The red lines represent
the multi-temperature spectra that best fits the
RHESSI detector-4 data for the same time interval
folded through the SOXS instrument response
matrix.
Summary
Detector Resolution

• The RHESSI results allow uncertainties to be
  determined on the fit parameters based on the
  differences between detectors.
• Comparisons between RHESSI and SOXS multi-thermal
  spectra at the peak of the 31 October 2004 flare
  show that their sensitivities are matched to
  within 10. Larger differences exist on the
  decay of the flare, where there is a difference
  of an order of magnitude in the emission measure
  at 2 keV. This is probably because of the lower
  temperatures and the lower energy coverage of
  SOXS and GOES compared to RHESSI.
• The measured 6.7 keV iron-line intensity suggests
  that the iron abundance in this flare is as low
  as 0.5 times the coronal abundance assumed by
  Chianti (v. 5.2), which is itself four times the
  photospheric abundance.
• Refinements to the RHESSI detector resolution (as
  small as 0.5 x nominal) and energy offsets (as
  large as 0.3 keV) are possible based on the
  measurements of the iron line.

response matrix.
Energy Offset