COCONUT-MF: Two-fluid ion-neutral global coronal modelling

Abstract

Context. The global coronal model COCONUT (COolfluid COronal uNstrUcTured) was originally developed to replace semi-empirical models such as the Wang-Sheeley-Arge model in space weather forecasting chains in order to improve the physical accuracy of the predictions. This model has, however, several simplifications implemented in its formulation to allow for rapid convergence in an operational setting. These simplifications include the assumptions that the plasma is fully ionised, sufficiently collisional, and that quasi-neutrality holds, so that it can be modelled as a single fluid. This means that all interactions with the low-concentration neutral fluid in the corona, such as collisions or charge exchange, are neglected. Aims. In this paper, we have two goals. Firstly, we aim to introduce a novel multi-fluid global coronal model and validate it with simple cases (like a magnetic dipole) as well as with real data-driven applications. Secondly, we aim to investigate to what extent considering a single-fluid plasma in the global coronal model might affect the resulting plasma dynamics, and thus whether the assumptions on which the single-fluid coronal model is based are justified. Methods. We developed a multi-fluid global coronal model following the ideal magnetohydrodynamics (MHD) COCONUT model, COCONUT-MF, which resolves the ion and neutral fluid equations separately. While this model is still steady-state and thus does not resolve unsteady processes, it can account for resistivity, charge exchange, and chemical (ionisation and recombination) and collisional contributions due to the presence of the neutrals in the fluid equations. Results. We present the results of the ion-neutral COCONUT-MF modelling for a magnetic dipole, a minimum of solar activity case (August 1, 2008), and a solar maximum case (March 9, 2016). Through comparison with the ideal MHD results, we confirm that the resolved multi-fluid solver features are physical and also demonstrate the higher accuracy of the applied upwind numerical flux scheme compared to the one used in the original MHD model. Subsequently, we also repeat the multi-fluid simulations while excluding the charge exchange and the chemical and collisional terms to evaluate the effect these terms have on the resulting plasma dynamics. It is observed in numerical results that, despite the very low concentration of neutrals, they still do affect the flow field to a limited but non-negligible extent (up to 5 to 10% locally), with a higher impact being seen in the case of the solar maximum. It is also demonstrated that the collisional terms are primarily responsible for the neutrals adopting the electromagnetic profiles of the ions, while the charge exchange and chemical terms yield the largest thermal effects of the neutrals on the ion plasma. Despite the fact that the coronal plasma is generally assumed to be collisionless, our results show that there is sufficient collisionality in it to couple the two fluids. Conclusions. We present a novel multi-fluid global coronal model that can separately simulate the behaviour of the ion and neutral fluids. Using this model, we also show that in our set-up, in which the chromosphere is not considered and steady-state solutions are assumed, the presence of the neutrals affects the flow field, though to a limited extent. It is shown that this effect is larger when the flow field is more complex due to a higher magnetic activity. This analysis may change in the future when the global coronal model will be extended to include the lower atmospheric layers as well as terms to model coronal heating, radiation, and thermal conduction. To that end, the current model may need to be further calibrated to better represent the different layers of the atmosphere. We presume that the use of the proposed COCONUT-MF set-up will then be necessary and new numerical experiments will need to be performed in order to confirm this hypothesis.