“…In order to circumvent the difficulty of numerically integrating the MHD equations in the presence of strong spatial gradients (without drastically reducing the spatial resolution of the numerical grid), Tanaka () proposed to solve the MHD equations by splitting the global magnetic field B into an intrinsic potential magnetic field B 0 and a residual magnetic field B 1 . This technique has been widely applied to MHD simulations of planetary magnetospheres since then (e.g., for recent studies of Saturn, Jia et al, ; Jupiter, Chané et al, ; or Uranus, Cao & Paty, ). However, in Tanaka (), the background magnetic field B 0 was assumed to be potential (i.e., current‐free) and time independent.…”