62. Coronal loop contraction and oscillation in flares

Author: Alexander Russell at the University of Dundee and Paulo Simões and Lyndsay Fletcher at the University of Glasgow.

<< previous nuggetnext nugget >>


Solar flares strongly perturb the corona and it is common to see coronal loops oscillate or contract following a flare. The two types of motion have traditionally been regarded as separate phenomena, but we suggest that they can be aspects of a single response to removal of magnetic energy from the corona. Interesting implications are that loop motions can be used as a diagnostic for the removal of coronal magnetic energy by flares, and that rapid change of coronal magnetic energy is a newly identified excitation mechanism for kink oscillations. A reassessment of previous observations, including a familiar event captured by TRACE in 1998, suggests that oscillating-imploding loops are a common feature in flares.


The M6.4 flare on 9th March 2012 produced a pronounced coronal implosion [1,2], accompanied by beautiful loop oscillations lasting 5 cycles. SDO/AIA 171 Å observations of these motions are presented in Figs. 1 and 2. The shortest, lowest loops contract with negligible oscillation, while the longer, upper loops oscillate significantly both during and after the contraction.

Figure 1: Movie of coronal loop motion in response to the flare. The arrows indicate contracting loops, several of which oscillate during the collapse.
Figure 2: Top: Time distance plot of loop positions using an artificial slit aligned with the tips of the arrows in the movie. L4 in the movie corresponds to C2 in this plot. Bottom: Fermi GBM count rate in the HXR 45–100 keV band showing start and duration of the flare impulsive phase.

Connecting displacement and oscillation

Figure 3: Removal-of-support mechanism for coronal loop motions near flares. (a) Initially, magnetic forces balance. (b) A flare converts magnetic energy, decreasing the magnetic pressure at the flare site. Forces on nearby loops become unbalanced. (c) Loops move towards the flare site.

Could oscillation and displacement be connected? And why do some loops oscillate but not others? Reduction of coronal magnetic energy provides neat answers to both questions.

First, consider why loops contract. Before the flare, the corona is approximately force-free, with magnetic tension and magnetic pressure gradient forces in balance (Fig. 3a). Flares reduce the magnetic energy density B2/2μ0 in part of the corona by converting it to other forms. This corresponds to a local decrease in the magnetic pressure; thus, the forces on nearby loops become unbalanced (Fig. 3b). The result is an inward motion or “implosion” towards the energy release (Fig. 3c) [3,4].

What form will the inward motion take? Since coronal loops have inertia they do not respond instantly to changes in their environment. Instead, stable coronal loops behave like driven oscillators and three classes of motion are possible depending on how the driver period (i.e. duration of the flare impulsive phase) compares to the oscillation period. If the equilibrium changes during a small fraction of the loop’s period, the result is a straightforward oscillation about the new equilibrium (Fig. 4a). At the opposite extreme, when the driver is slowly applied over many periods, no oscillation is detected (Fig. 4b). In the intermediate regime where driving and oscillation timescales are broadly similar (Fig. 4c), the loop motion becomes an oscillation superimposed on the displacement that excited it.

Figure 4: Loop motions according to driver timescale and oscillation period. (a) A rapid change in equilibrium impulsively excites oscillations. (b) Comparatively slow changes produce a gradual displacement. (c) Oscillation during displacement occurs when the timescales are broadly similar.

Comparing the observations in Fig. 2 with the responses illustrated in Fig. 4, L3 is a near-perfect example of oscillation during displacement and L2 also belongs to this category. The uppermost loop, L1, appears to be at the borderline between this class and the impulsively excited oscillation. The lowest loops, C1 and C2, fit the pattern of gradual collapse. Since loop period increases with loop length from C2 to L1, the ordering of the types of motion agrees with the model.

For this event, the duration of the loop contractions in Fig. 2 implies that most of the energy conversion happened in the first 300 s; and the large amplitude oscillations imply a sharp switching-on of the energy release, during 40 s or less. We also conclude that the flare was powered by magnetic energy in the heart of the active region, below the contracting loops and close to the flaring arcade.

Oscillating imploding loops are common

Figure 5: TRACE 171 Å observation of imploding oscillating loops from 14th July 1998 (movie).

The motivating event is a particularly clear example, but we have found oscillating imploding loops for a broad range of X and M class flares.

As an example of historical interest, Fig. 5 shows the famous 14th July 1998 TRACE observation that confirmed the presence of standing kink oscillations in the corona. Loops in the bottom right hand corner oscillate seemingly due to an outward impact such as from a blast wave [5]; however, the loops indicated by the arrow contract inwards and this seems to drive at least one oscillation. The initial movement and irreversible displacement are consistent with a removal of support mechanism (time-distance plots for these loops can be found in [6]).

Another interesting event is the 8th May 2012 M1.4 flare studied by [7], where loop motions suggest a very localised decrease in coronal magnetic energy due to the confined flare.


Rapid conversion of coronal magnetic energy makes coronal loops move towards the volume where magnetic energy has decreased, and because loops have inertia, oscillations are an intrinsic part of the motion. Loop motions have potential to reveal the changes in coronal magnetic energy that have long been thought to power flares, while displacement of loops is a newly identified excitation mechanism for kink oscillations. For further details see [3].


  • [1] Simões, P. J., Fletcher, L., Hudson, H. S. and Russell, A. J. B. 2013, UKSP Nugget 42
  • [2] Simões, P. J., Fletcher, L., Hudson, H. S. and Russell, A. J. B. 2013, ApJ 777, 152
  • [3] Russell, A. J. B., Simões, P. J. and Fletcher, L. 2015, A&A 581, A8
  • [4] Hudson, H. S. 2000, ApJ 531, L75
  • [5] Nakariakov, V. M., Ofman, L., Deluca, E. E., Roberts, B., & Davila, J. M. 1999, Sci, 285, 862
  • [6] Aschwanden, M. J., Fletcher, L., Schrijver, C. J., & Alexander, D. 1999, ApJ, 520, 880
  • [7] White, R. S., Verwichte, E. and Foullon, C. 2013, ApJ 774, 104