V.I. Abramenko, V.B. Yurchyshyn, H. Wang , T.R. Spirock, P.R. Goode

1

• V.I. Abramenko, V.B. Yurchyshyn, H. Wang , T.R.
  Spirock, P.R. Goode
• Big Bear Solar Observatory, NJIT
• Crimean Astrophysical Observatory, Ukraine
• Email avi_at_bbso.njit.edu

34th Meeting of SPD 16-29 June 2003
2

INTRODUCTION
Analysis of the non-thermal broadening of soft
X-ray spectral lines in solar flares observed
with Yohkoh (Alexander et al. 1998, Harra et
al. 2001) showed that the non-thermal velocity
begins to rise before the flare onset and peaks
often before the Hard X-ray emission.
COES X-ray flux
The non-thermal velocity
?
? 11 min - the growth time of the non-thermal
velocity
There are changes in the turbulent state of an
active region, leading to the flare onset, in
other words, there is a preflare turbulent phase.
3
INTRODUCTION
Due to the magnetic coupling between the
corona and the photosphere (Parker 1979, 1996),
preflare turbulent phase may involve the
photosphere, too.
Photospheric plasma is in a state of highly
developed turbulence, where the vertical
component of the magnetic field, Bz, diffuses in
the same way as a passive scalar in a turbulent
flow (Parker 1979, Petrovay and Szakaly 1993).
Thus, we can apply methods of the theory of
turbulence to the longitudinal magnetic field
of an active region measured near the center of
the solar disk.
4
OBSERVATIONAL DATA
B
Longitudinal magnetic field
??
from Big Bear Solar observatory Video
(upper penal) and Digital (lower
penal) Magnetograph Systems
The M8.4 flare
The X9.4 flare
Pixel sise 0.6 x 0.6 arcsec
Measurements covered the time periods before,
during and after a major flare with an
appropriate time cadence.
The X1.6 flare
The M8.7 flare
5
METHOD
A. The degree of intermittency of the magnetic
field

An increase in the turbulence implies that the
turbulence becomes more intermittent.
Intermittency characterizes a tendency of a
turbulent field to concentrate into widely
spaced very intense small-scale features.
Frisch, 1995
An example of highly intermittent structure

6
METHOD
A. The degree of intermittency of the magnetic
field
The degree of intermittency may be estimated by
determining structure functions of high
statistical orders

Here, q is the order of a statistical moment, r
is a separation vector, x is the current point
on a magnetogram. ltgt denotes the averaging over
a magnetogram. ??q? is a slope within the
inertial range of scales.
Non-intermittent turbulence

The routine was proposed by Abramenko et
al. ApJ 577, 2002
7
METHOD
A. The degree of intermittency of the magnetic
field
Non-intermittent turbulence


??-1
Highly intermittent turbulence

???
8
METHOD
B. Correlation length of the magnetic energy
dissipation field
For the longitudinal component of the
photospheric magnetic field the energy
dissipation, per unit mass in a unit of time, can
be written (Monin Yaglom 1975)
For every magnetogram we calculated the magnetic
energy dissipation structure, ??x,y?. The
correlation length of these these clusters, ?,
was determined using the method of the
turbulence theory (Monin and Yaglom 1975).
9

RESULTS
10
CONCLUSIONS
Our results - support the existence of
the preflare turbulent phase in an active region
(Alexander et al. 1998, Harra et al. 2001)
- are in agreement with the concept that a
solar flare is the collective energy released by
an avalanche of reconnection events at
small-scale discontinuities of the magnetic
field (the self-organized criticality concept )
(Parker 1987 Longcope and Noonan 2000
Charbonneau, McIntosh, Liu and Bogdan
2001) - show that statistical properties
of a flare-related nonlinear dissipative process
in an active region can be studied by using the
photospheric longitudinal magnetic field.
11

The X1.6 flare
The X9.4 flare
12
The M8.7 flare
13
First, we calculated the correlation function
B(r ) ? (?(xr) - ? ? ?)(?(x)- ? ? ?)?
We have to normalize B(r) by the variance of
dissipation
b(r) B(r) / B(0)
By integrating b(r), over all scales r, we obtain
a correlation length of the energy dissipation
structure
max
r
? ? b(r) dr
?
Correlation length of the magnetic energy
dissipation cluster
14
The M8.4 flare on Nov 5, 1998 in active region
NOAA 8375
GOES
H?
c
Flux
?
?
15
The M8.7 flare on July 26, 2002 in active region
NOAA 0039
GOES
H?
Flux
c
16
The X1.6 flare on October 19, 2001 in active
region NOAA 9661
GOES
H?
Flux
c
?
?
17
The X9.4 flare on March 22, 1991 in active region
NOAA 6555
GOES
Flux
c
?
?
18
Table 1.
19
Table 2.
20
CONCLUSIONS
1.In all of the cases we found a peak in ?, which
was followed by a peak in ?. During the time
interval between them, ? , a rapid growth of the
soft X-ray and H? flux occurred.
?
2.The peak in beta was preceded by a period of
gradual growth of ?, ? . Maximum in ? occurred
earlier than the peak of the hard X-ray emission.
?
3. The maximum of ? tends to follow or to occur
nearly simultaneously (with the accuracy of about
2-5 min) with the maximum of the Hard X-ray
emission.
4. Based on limited examples, we conclude that
the time intervals ? and ? are inversely
proportional to impulsivity and intensity of
flares.
?
?