Secondary small-scale dynamics of a Rayleigh-Taylor unstable solar prominence

Aims. Quiescent solar prominences clearly show small-scale dynamics in observations. Their internal properties realise density contrasts with the Sun’s atmosphere that must be liable to mainly Rayleigh-Taylor (RT) instabilities, which in turn lead to the formation of a vertically dominated prominence structure when viewed at the solar limb. As a result, prominences develop bubbles and plumes but also secondary instabilities in the form of Kelvin-Helmholtz (KH) roll-ups along the edges of the bubbles and plumes. Recent observations also indicate the existence of reconnection events within the RT turbulent flows. Methods. We ran 2.5D high-resolution resistive magnetohydrodynamic simulations with the open-source <ns0:monospace>MPI-AMRVAC</ns0:monospace> code. We achieved a spatial resolution of ∼11.7 km in a 2D domain of size 30 × 30 Mm and ran the simulation for about 10 minutes of solar time. A dense, magnetic pressure-supported prominence at coronal heights acts as our initial state which was then perturbed by the RT instability at the prominence corona interface. The localised current concentrations which are formed as a result of multiple, interacting instabilities were studied statistically. Results. The combination of primary (RT) and secondary (KH) instabilities leads to the formation of current sheets and causes an interplay between turbulent flow patterns and magnetic reconnection. The outflows from the reconnection regions lead to the formation of energetic jets from the current sheets facilitating energy exchange and dissipation throughout the prominence structure. We analysed our high-resolution prominence simulation using synthetic images of the broadband SDO/AIA 094, 171, and 193 Å and narrowband Hα filters to compare the developing fine-scale structures with their observational counterparts. We find the majority of the secondary instabilities to form in the hotter regions surrounding the cooler prominence material. Conclusions. The secondary instabilities and current sheets in our simulation agree with observations (scale, speed, and duration) yet the simulated activity is predominantly localised in hot, surrounding coronal plasma rather than the cool prominence material. This inconsistency points to missing ingredients rather than a shortcoming of the KH interpretation, and so 3D follow-up studies should revisit these findings in more realistic magnetic topologies.