Highlights • MT index can be used to estimate the contribution of magnetotail current to the Dst index. • WINDMI can successfully model the magnetotail contribution to the ground magnetic perturbation. • During substorms the intensity of the geomagnetic tail drops and is reflected in corresponding decrease in the MT index latitudes. • WINDMI model can reliably estimate the change in geotail current during substorms.

Abstract In this paper, we have improved the capabilities of a low dimensional nonlinear dynamical model called WINDMI to determine the state of the global magnetosphere by employing the magnetotail (MT) index as a measurement constraint during large geomagnetic storms. The MT index is derived from particle precipitation measurements made by the NOAA/POES satellites. This index indicates the location of the nightside ion isotropic boundary, which is then used as a proxy for the strength of the magnetotail current in the magnetosphere. In Asikainen et al. (2010), the contribution of the tail current to the Dst index is estimated from an empirical relationship based on the MT index. Here the WINDMI model is used as a substitute to arrive at the tail current and ring current contribution to the Dst index, for comparison purposes. We run the WINDMI model on 7 large geomagnetic storms, while optimizing the model state variables against the Dst index, the MT index, and the AL index simultaneously. Our results show that the contribution from the geotail current produced by the WINDMI model and the MT index are strongly correlated, except during some periods when storm time substorms are observed. The inclusion of the MT index as an optimization constraint in turn increases our confidence that the ring current contribution to the Dst index calculated by the WINDMI model is correct during large geomagnetic storms.