Yohkoh statistical studies

1
Yohkoh statistical studies

• Michal Tomczak
• Astronomical Institute,
• University of Wroclaw, Poland

2
Introduction

• Yohkoh Hard X-ray Telescope gave for the first
  time an opportunity for massive investigation of
  spatial distribution of hard X-ray emission in
  solar flares the mission-long database (Oct 1,
  1991 Dec 14, 2001) contains 3071 events.

3
Observational constraints

• to obtain a valuable hard X-ray image about
  100200 cts/SC are needed ? flux limit.
• to distinguish different sources a flare size
  should be larger than spatial resolution of the
  instrument (we cannot resolve events having h lt
  89 x 103 km) ? size limit.
• to distinguish coronal and footpoint sources a
  contamination of their photons should be omitted
  event should be seen on-side ? geometry limit.

4
Selection criteria (Masuda 1994)

• peak count rate in the M2band exceeding 10 cts
  s-1 SC-1 at least one image is available in the
  channel which records the radiation of purely
  non-thermal electrons.
• heliocentric longitude exceeding 80º.

5
Comparison between surveys
Survey Period Total number of flares Number of selected events
Masuda (1994), Ph. D. thesis Oct 91 Sep 93 850 10
Petrosian et al. (2002), ApJ, 569, 459 Oct 91 Aug 98 1307 18
Tomczak Ciborski (2006), AA preprint Oct 91 Dec 01 3071 117
6
What have the surveys taught us about coronal
sources?

• They are a common feature of solar flares (M
  7/10 P 15/18 TC 45/117).
• During the impulsive phase they are usually
  fainter than footpoint sources, a disproportion
  becomes more important at bursts maxima and for
  higher photon energies.
• Their photon energy spectra are usually softer
  than the spectra of footpoint sources (an
  exception the above-the-loop-top sources).

7
L
M1
Masuda 1994
8
August 18, 1998
Petrosian et al. 2002
9
Petrosian et al. 2002
Tomczak Ciborski 2006
10
Masuda 1994
Footpoint sources
Loop-top sources
11
Petrosian et al. 2002
?FT 4.9 1.5 ?LT 6.2 1.5
12
HXR imaging constraints

• The finite dynamic range of the HXT estimated to
  be about 1 decade (Sakao 1994).
• The generation of spurious sources by the
  reconstruction routines ? a false photometry of
  real sources.
• Weak sources suppression in the presence of
  strong sources (Alexander Metcalf 1997) ? the
  reconstruction routines treat them as a
  statistically insignificant.

13
Alexander Metcalf 1997, ApJ, 489, 442
14
To study weak sources in the presence of strong
sources is our case! HXR imaging constraints
limit a possibility of investigation of coronal
sources. Moreover, our results can be even false
due to the light curve mimicking that of the
footpoints.
15
What can we do?

• To wait for modern instruments with a better
  dynamic range as well as for clever
  reconstructions routines.
• To investigate only examples in which the coronal
  sources dominate ? we obtain only a partial
  picture.
• To eliminate somehow stronger footpoint sources

16
Behind-the-limb flares

• We use the solar limb as a screen which
  occults the lower part of the flaring structure
  (footpoint sources) and leaves emission of the
  higher part only (coronal sources)
• in this way we can separate coronal sources of
  all type our choice do not favour any particular
  physical mechanism,
• - we loose a possibility of the comparison with
  the footpoint sources.

17
The behind-the-limb configuration has been used
in many different way

• for HXRs (e.g. Frost Dennis 1971, stereo-scopic
  papers of Kane).
• Yohkoh Bragg Crystal Spectrometer diagnostics of
  bright SXR loop-top kernels (e.g. Khan et al.
  1995, Mariska et al. 1996, Mariska McTiernan
  1999).
• BCS diagnostics of X-ray plasma ejections
  (Tomczak 2005).

18
How can we find out such events?

• Compare the GOES list of flares to the Ha list
  from the SGD events present in the first one and
  absent in the second one can be behind-the-limb
  type ? a prompt selection.
• Check manually soft X-ray images of the
  prompt-selected flare ? impulsive SXR
  brightenings should not be seen
• Check a time of the limb passage for the active
  region in which the prompt-selected flare has
  occurred ? ?(t) extrapolation

19
Tomczak (2001), AA, 366, 294

• 14 behind-the-limb flares that occurred between
  19911994 has been selected.
• In this case to investigate time variation of
  coronal sources we need not actually HXR images!
  ? our temporal resolution becomes better.

20
(No Transcript)
21
(No Transcript)
22

• Additional 16 partially occulted flares observed
  by Yohkoh in years 19971999 are preparing
  (Tomczak Sokolnicki).

23
Masuda flare

• Masuda et al.1994, Nature, 371, 495 about 300
  citations in the Smithsonian/NASA Astrophysics
  Data System.
• Tomczak Ciborski (2006) an additional argument
  confirming how unusual this event is.

24

25
Explanation

• Different converging field geometry events from
  branch A occurred in more converged loops than
  the events from branch B the more converged
  loops from branch A correspond to flux tubes that
  are less helically twisted the less converged
  loops from branch B are more twisted.
• Anomalous electron scattering absent for events
  from branch A, present for events from branch B.