Modelling the propagation of flux ropes in the coronal model COCONUT

Abstract

Some space weather forecasting tools, such as the EUropean Heliosphere FORecasting Information Asset (EUHFORIA), consist of two parts. The first is a semi-empirical coronal model used to create the background solar wind, and the second is a heliospheric model in which coronal mass ejections (CMEs) are injected through the inner boundary located at 0.1 AU. In these models, the inserted CME does not interact with the solar wind before the inner boundary. To take this interaction into account and provide a more realistic description of CMEs at 0.1 AU. We studied the propagation of flux ropes from the solar surface to 0.1 AU in a full magnetohydrodynamic coronal model called COCONUT (COolfluid COroNal UnsTructured). The CMEs were modeled using the modified Titov-Démoulin model (TDm) of flux rope. We tracked the evolution of twenty four different twisted flux ropes within realistic corona configurations reconstructed by COCONUT from the GONG magnetic maps of both minimum and maximum solar activity. All CMEs are identical except for their net initial current. Our results reflect dynamic expected by the standard flare model, such as presence of post-flare loops and the pinching of the CME's legs. However, the shape of the CME varies greatly depending on whether the solar wind corresponds to a minimum or a maximum activity, highlighting the crucial role of the solar wind in determining the geometry of CMEs. Once the flux ropes reach 0.1 AU, their thermodynamic and magnetic properties are extracted. We found that, for the two solar wind configurations, the synthetic profiles obtained are consistent with those that satellites could measured. Moreover, simple relationships are emphasised between the net initial current of flux ropes and the shape of the different synthetic profiles.  Finally, using this CME description, the boundary conditions imposed on EUHFORIA (or other heliospheric models) should be more accurate than those provided by an independent CME model and therefore lead to more realistic forecasts.​