Linear separation of orthogonal merging component and viscous interactions in solar wind-geospace coupling

The general view is that the response of the magnetosphere-ionosphere system to a complex, variable solar wind input will be a complicated and nonlinear function of that input. We investigate this question using a global MHD code to simulate the interaction and determine the responses of the system to isolated aspects of the solar wind input. We present evidence from the simulations that the solar wind-geospace interaction can be linearly decomposed into component merging interactions for interplanetary magnetic field (IMF) B_y separate from B_z and a separate viscous interaction. One can run the global simulation (in our case, the Lyon-Fedder-Mobarry code) using just the solar wind plasma and B_z time series, do a second simulation using just the solar wind plasma and B_y time series, add the results of the two simulation outputs, then subtract the output from a simulation done with only the plasma input and no magnetic field, since the sum of the B_y and B_z runs has two viscous interactions (one for each run), and get an output that is very close to the result of a single run using the entire IMF merging field (B_y and B_z) along with the plasma time series. This demonstrates that the components of merging and viscous interactions between the solar wind and geospace are linearly separable to a very large degree.