Effects of the low‐latitude ionospheric boundary condition on the global magnetosphere

In common treatment of magnetosphere‐ionosphere coupling at high latitudes, the ionosphere is represented by a thin conducting spherical shell, which closes field‐aligned currents generated in the magnetosphere. In this approach, the current continuity yields a Poisson equation for the electrostatic potential associated with the ionospheric convection pattern. Solution of the Poisson equation then provides a means of self‐consistently describing magnetospheric and ionospheric plasma convection with a feedback of one on the other. While the high‐latitude ionospheric convection is driven by the solar wind and magnetosphere interaction, at lower latitudes atmospheric neutral winds start to dominate. The question that arises then is whether and how midlaltitude and low‐latitude ionospheric convection affects high‐latitude ionospheric and magnetospheric convection. In global magnetospheric models, ionospheric convection equatorward of the low‐latitude boundary is excluded from the simulation domain. However, the boundary condition applied at that boundary to the electrostatic potential may be used as a proxy of this convection. In this paper, we explore effects that different idealized low‐latitude boundary conditions have on the magnetospheric configuration simulated by the Lyon‐Fedder‐Mobarry global magnetohydrodynamic model. To this end, we perform a number of idealized simulations different only in the low‐latitude ionospheric boundary condition used. We find that the behavior of the system can be influenced rather significantly by the different boundary conditions, which is expressed by changes in the evolution of the polar cap potential, global magnetospheric convection, and plasma pressure distribution in the magnetotail and on the dayside. The differences in the cross‐polar cap potential can reach up to >10%, dependent on the boundary condition used. In the magnetosphere the low‐latitude ionospheric boundary condition affects the strength and location of the plasma outflow from the distant tail x‐line and the subsequent earthward convection. Changes in the plasma pressure distribution on the nightside are accompanied by noticeable differences in the shape of the magnetotail. We confirm that the changes in the magnetospheric and ionospheric configuration are not just temporal deviations of the system from the same average dynamical state by considering 1 h averages of the magnetospheric flow and pressure distribution. These results verify that the simulated system reaches similar but distinctly different dynamical states dependent on the low‐latitude boundary condition applied.