RHESSI observations of LDE flares – extremely long persisting HXR sources

Mrozek, T., Kołomański, S., Bąk-Stęślicka, U.

Astronomical InstituteUniversity of Wrocław

At last!

YOHKOH results - SXR

Long Duration Event (LDE)Long Duration Flare (LDF)Long Duration Arcade Flare (LDAF)

Kołomański, S., 2007:

> 6h duration

> 3 orbits of YOHKOH starting from the maximumof the flare

YOHKOH results - SXR

Different sources observed at the same time suggest that the energy reales takes placein different places

Typical size of the SXR source (LDE case): 1.0-1.5x104 km

YOHKOH results - HXR

HXR emission in the L channel (14-23 keV) was observed up to 40 minutesafter the maximum of the flare.

1 2

21

YOHKOH results - HXR

Rise phase – coronal and footpoint sources

Decay phase - HXR source observed 40 minutes after the maximum of the flare. It is 10 times longer than characteristic cooling time of such source – indirect proof for the energy release long after the maximum of the flare.

RHESSI & LDEs - motivation

Better spatial resolution – more detailed investigation of sources

Better sensitivity - weak, coronal sources could be detected long after the maximum of the flare

Better energy resolution – more detailed analysis of LDEs spectra

Difficulties

Main difficulties:

- pile-up - attenuators- orbital background

RHESSI & LDE

Feb. 2002 – Feb. 2008

~ 160 LDE flares foundwith the use of GOESlightcurves

~ 50 which last longer than 3 hours in RHESSI observations(6-12 keV)

30 July 2005X1.3>10 h

Method

2-minutes intervals:

- attenuators out

- outside the radiation belts

- far from the SAA

Thus, for 10 hours decay we have only few time intervals for imaging and spectroscopy

Method

Images:

Time interval: 11:38 – 11:40Grids: 3,4,5,6,8,9Pixel size: 1”

Method

Images:

Time interval: 11:38 – 11:40Grids: 3,4,5,6,8,9Pixel size: 1”

4-6 keV 10-12 keV 15-23 keV

Method

The signal in the 12-25 keV interval is observed (11:40 UT – 6 hoursafter the maximum) - why we can’t obtain images?

Method

The signal in the 12-25 keV interval is observed (11:40 UT – 6 hoursafter the maximum) - why we can’t obtain images?

Because of the actual size of the source? – let’s look at the single-detector images

Method

The size of sources changes

When the diameterof the source is larger than the FWHM of given grid then the modulation vanishesand the source is no longer observed with this grid.

For this reason we have to choose grids in more flexible way

gridnumber

time

Method

As the result we obtain well resolved sources.

Time interval: 11:38 – 11:40Grids #: 8,9 Algorithm: PIXONEnergy ranges [keV]:

5-6, 7-8, 9-10, 11-12, 12-14, 15-23

FWHM of the grid #8 is about 100 arc sec

30 July 2005 - images

Comparison with EIT 195 Å

RHESSI images reconstructed with the use of PIXON method

Red contours – 6-7 keVBlue contours – 15-25 keV

6 hours after the maximum of the flare

What is the nature of this source?

30 July 2005 - spectra

double thermal

EM: 9.3x1047cm-3, T: 9.3 MK

EM: 6.4x1045cm-3, T: 20 MK

30 July 2005 - spectra

thermal + thin target

EM: 7.7x1047cm-3, T: 9.9 MK

δBB: 7.4, EB: 100 keV, δAB: 20, Ecut: 11.2 keV

30 July 2005 - spectra

thermal + thick target

EM: 7.3x1047cm-3, T: 10.0 MK

Fe: 3.4x1034 s-1, δBB: 12.2, EB: 600 keV, δAB: 6.0, Ecut: 13.9 keV

30 July 2005 - spectra

thermal + broken power-law

EM: 6.5x1047cm-3, T: 10.0 MK

γ BB: 1.7, EB: 12.0 keV, γAB: 10.0

30 July 2005

Having the temperature, emission measure, size and height we were able to estimate the energy balance.

To balance the thermal and conductive losses we need a heating of the order of 1 erg s-1cm-3

(1028 erg s-1 from the whole volume)

Typical size of the long persisting HXR source is of the order of 104 km

Models, models…

Shibata 1995

The main driver of the whole process is the eruption of the filament

The several hours long energy release cannot be explained with this scenario.

Jakimiec, J., et al. 1998

The existence of the turbulent (highly tangled) magnetic field in the loop-top source could keep up the energy release for long time due to small reconnections inside the structure.

It explains the spatial correlation between the thermal and non-thermal/hot sources observed in the late phase of LDEs

Sweet, P. A. 1958

Emergence of the new flux is the main driver in this model.

This idea was recently resurrected by Uchida et al. (1999) and Hirose et al. (2001)

Conclusions

LDEs are well observed by RHESSI. The analysis is complicated due to attenuators, radiation belts, SAA, but not impossible.

HXR sources (above 15 keV) are visible even 6 hours after the maximum of the flare.

Long-lasting HXR sources are located above structures observed in the EUV range. Observed sources are large and grows with time.

The spectral analysis of the sources suggests that there are at least two components present. One is the hot (about 10 MK) and the second is a very hot (20 MK) or steep non-thermal component.

The observed features imply the existence of the energy release process which lasts several hours.