Analysis of the 3–7 October 2000 and 15–24 April 2002 geomagnetic storms with an optimized nonlinear dynamical model

Abstract

A computationally optimized low‐dimensional nonlinear dynamical model of the magnetosphere‐ionosphere system called WINDMI is used to analyze two large geomagnetic storm events, 3–7 October 2000 and 15–24 April 2002. These two important storms share common features such as the passage of magnetic clouds, shock events from coronal mass ejections, triggered substorms, and intervals of sawtooth oscillations. The sawtooth oscillations resemble periodic substorms but occur in association with strong or building ring current populations and have injection regions that are unusually close to the Earth and unusually wide in magnetic local times (Henderson et al., 2006; Borovsky et al., 2007). The April 2002 event includes one of the best examples of sawtooth events ever observed. On 18 April 2002, sawtooth oscillations were clearly visible when solar wind conditions (IMF Bz, density, pressure) were relatively steady with a slowly varying Dst. In this study, WINDMI is used to model the 3–7 October 2000 and 15–24 April 2002 geomagnetic activity. WINDMI results are evaluated focusing on the sawtooth intervals and the overall prediction of the westward auroral electrojet (AL) index and Dst index. The input to the model is the dynamo driving voltage derived from the fluctuating solar wind plasma and the interplanetary magnetic field measured by the ACE satellite. The output of the model is a field‐aligned current proportional to the AL index and the energy stored in the ring current which is proportional to the Dst index. The model parameters are optimized using a genetic algorithm (GA) to obtain solutions that simultaneously have least mean square fit to the AL and Dst indices and also exhibit substorms of period 2–4 hours. The GA optimization results show that the model is able to predict the Dst index reliably and captures the timing and periodicity of the sawtooth signatures in the AL index reasonably well for both storm events.

The effect of nonlinear ionospheric conductivity enhancement on magnetospheric substorms

“…The onset of the substorm was between 03:05 and 03:10 UTC on the day . With the solar wind parameters obtained from the ACE spacecraft translated into the nose of the magnetosphere (Spencer et al, 2007), the resultant input rectified voltage is shown in the bottom panel of Fig. 3.…”

“…The WINDMI model is described in some detail in , Spencer et al (2007) and more recently in Patra et al (2011). The equations of the model are given by the following:…”

“…In this work, we modify the equations of a low-order, physics-based nonlinear model of the magnetosphere called WINDMI Spencer et al, 2007), in order to account for the contribution of ionospheric conductivity enhancement to substorm behavior. The current version of the model is available at the NASA Community Coordinated Modeling Center for near real time forecasts of space weather activity (Mays et al, 2009).…”

Evaluation of solar wind‐magnetosphere coupling functions during geomagnetic storms with the WINDMI model

“…The plasma physics‐based WINDMI model uses the solar wind dynamo voltage V sw generated by a particular solar wind‐magnetosphere coupling function to drive eight ordinary differential equations describing the transfer of power through the geomagnetic tail, the ionosphere and the ring current. The WINDMI model is described in some detail by Doxas et al [2004], Horton et al [2005] and more recently by Spencer et al [2007]. The equations of the model are given by The largest energy reservoirs in the magnetosphere‐ionosphere system are the plasma ring current energy W rc and the geotail lobe magnetic energy W m formed by the two large solenoidal current flows ( I ) producing the lobe magnetic fields.…”

“…However an optimized linear scale yields better results of λ AL than the fixed scale which does not take into account changes in width, height, and location of the electrojet during geomagnetic activity. The scaling factor for the 3–7 October 2000 storm was calculated to be 3275, while for the 15–24 April 2002 storm it was computed to be 2638, both in A / nT [ Spencer et al , 2007].…”