On the Propagation of a Geoeffective Coronal Mass Ejection during March 15 -- 17, 2015

The largest geomagnetic storm so far in the solar cycle 24 was produced by a fast coronal mass ejection (CME) originating on 2015 March 15. It was an initially west-oriented CME and expected to only cause a weak geomagnetic disturbance. Why did this CME finally cause such a large geomagnetic storm? We try to find some clues by investigating its propagation from the Sun to 1 AU. First, we reconstruct the CME's kinematic properties in the corona from the SOHO and SDO imaging data with the aid of the graduated cylindrical shell (GCS) model. It is suggested that the CME propagated to the west $\sim$$33^\circ$$\pm$$10^\circ$ away from the Sun-Earth line with a speed of about 817 km s$^{-1}$ before leaving the field of view of the SOHO/LASCO C3 camera. A magnetic cloud (MC) corresponding to this CME was measured in-situ by the Wind spacecraft two days later. By applying two MC reconstruction methods, we infer the configuration of the MC as well as some kinematic information, which implies that the CME possibly experienced an eastward deflection on its way to 1 AU. However, due to the lack of observations from the STEREO spacecraft, the CME's kinematic evolution in interplanetary space is not clear. In order to fill this gap, we utilize numerical MHD simulation, drag-based CME propagation model (DBM) and the model for CME deflection in interplanetary space (DIPS) to recover the propagation process, especially the trajectory, of the CME from $30 R_S$ to 1 AU. It is suggested that the trajectory of the CME was deflected toward the Earth by about $12^\circ$, consistent with the implication from the MC reconstruction at 1 AU. This eastward deflection probably contributed to the CME's unexpected geoeffectiveness by pushing the center of the initially west-oriented CME closer to the Earth.