Ensemble modeling of successive halo CMEs observed during 2-4 Aug 2011

Abstract

Previously, we investigated the sensitivity of the 3D MHD Wang-Sheeley-Arge (WSA)-Enlil numerical results to the input “cone” geometry for an Earth-directed “halo” coronal mass ejection (CME) event [1]. A modeling ensemble was created from multiple sets of input parameters obtained through the use of the cone fitting tool together with realistic ranges for the angular width and leading edge distance estimated from STEREO limb observations. We obtained a spread of ~ 13 hours in the ensemble shock arrival times, which overlapped and was nearly centered on the observed arrival time. Because the sensitivity of the modeling ensemble may be event-dependent, we conduct an ensemble modeling study for three consecutive halo CME events. These events were selected in part because of the availability of STEREO limb observations, which help to constrain the initial CME geometries during the cone fitting process, but also because of the opportunity to explore the predictive capability of WSA-Enlil-cone in modeling multiple and interacting CMEs. Our preliminary results show that all three CMEs need to be modeled together in order to capture the gross features of the disturbed solar wind structure, including the shock arrival times, observed at 1 AU. For one multi-CME ensemble simulation, we obtain an “average” arrival time prediction that is ∼4 hours after the observed arrival time and an ensemble spread in the arrival time of ∼4 hours.