Forward modelling of MHD waves in braided magnetic fields

Abstract

Aims. We investigate synthetic observational signatures generated from numerical models of transverse waves propagating in complex (braided) magnetic fields. Methods. We consider two simulations with different levels of magnetic field braiding and impose periodic, transverse velocity perturbations at the lower boundary. As the waves reflect off the top boundary, a complex pattern of wave interference occurs. We applied the forward modelling code FoMo and analysed the synthetic emission data. We examined the line intensity, Doppler shifts, and kinetic energy along several line-of-sight (LOS) angles. Results. The Doppler shift perturbations clearly show the presence of the transverse (Alfvénic) waves. However, in the total intensity, and running difference, the waves are less easily observed for more complex magnetic fields and may be indistinguishable from background noise. Depending on the LOS angle, the observable signatures of the waves reflect some of the magnetic field braiding, particularly when multiple emission lines are available, although it is not possible to deduce the actual level of complexity. In the more braided simulation, signatures of phase mixing can be identified. We highlight possible ambiguities in the interpretation of the wave modes based on the synthetic emission signatures. Conclusions. Most of the observables discussed in this article behave in the manner expected, given knowledge of the evolution of the parameters in the 3D simulations. Nevertheless, some intriguing observational signatures are present. Identifying regions of magnetic field complexity is somewhat possible when waves are present; although, even then, simultaneous spectroscopic imaging from different lines is important in order to identify these locations. Care needs to be taken when interpreting intensity and Doppler velocity signatures as torsional motions, as is done in our setup. These types of signatures are a consequence of the complex nature of the magnetic field, rather than real torsional waves. Finally, we investigate the kinetic energy, which was estimated from the Doppler velocities and is highly dependent on the polarisation of the wave, the complexity of the background field, and the LOS angles.

Magnetic reconnection and the Kelvin-Helmholtz instability in the solar corona

“…This results in a fraction of the injected energy exciting a slow wave (see, for example, Fig. 10 in Fyfe et al 2020). It is theoretically possible to impose a wave driver that satisfies v •B = 0 everywhere on the lower z boundary, however, this would require a precisely varying profile.…”

Investigating coronal wave energy estimates using synthetic non-thermal line widths

“…However, in this case, the magnetic field structure is a lot more complex (complex magnetic field model). The simulation used here was previously discussed and investigated by Howson et al (2020a) and subsequently forward modelled by Fyfe et al (2020).…”

“…For this model, the numerical domain consists of a 256 × 256 × 1024 grid, which covers physical dimensions of 30 Mm × 30 Mm × 100 Mm. However, within the forward modelling analysis in Fyfe et al (2020), which we subsequently used to obtain the non-thermal line widths (see Sect. 3), the grid used in Howson et al (2020a) was spatially resampled to every fourth grid cell along x, y, and z.…”

“…This was to reduce the computational cost and was shown to have no significant impact on the synthetic spectroscopic data. For more information on the behaviour and forward modelling of the simulation, we direct the reader to Howson et al (2020a) and Fyfe et al (2020), respectively.…”

“…Even though the driver is mono-periodic and linearly polarised along LOS y , comparing it to the simulation with the equivalent driver in PVD2020 and, hence, using a ratio threshold of √ 2 is not an appropriate comparison. This is firstly because we considered LOS x , which is not aligned along the direction of the driver, and secondly Howson et al (2020a) and Fyfe et al (2020) have shown that the polarisation of the waves changes from strictly v y at the driver to also containing a v x component throughout the rest of the 3D domain. Therefore, this simulation must be compared to the lowest threshold PVD2020 present; hence examining the ratios with respect to the threshold σ nt /v rms > 1 is used here (see Table 2).…”