We present a study of interplanetary coronal mass ejection (ICME) propagation to Earth and Mars. Because of the significant space weather hazard posed by ICMEs, understanding and predicting their arrival and impact at Mars is important for current and future robotic and manned missions to the planet. We compare running ENLILv2.6 with coronal mass ejection (CME) input parameters from both a manual and an automated method. We analyze shock events identified at Mars in Mars Global Surveyor data in 2001 and 2003, when Earth and Mars were separated by <80° in heliocentric longitude. The shocks identified at Mars were also identified at Earth, and the majority of the shock sources were identified through the Solar and Heliospheric Observatory–Large Angle and Spectrometric Coronagraph catalogue. We find that arrival times predicted by the two methods at both planets are statistically similar, dynamic pressures predicted when using the automated method are better, and the automated method tends to underestimate both CME width and speed. Using the location of the related flare as the CME direction did not improve results. In addition, changing the CME speed toward the plane‐of‐sky speed at 20
R Simproves the match to observations, mainly because the speed found by the automated method is underestimated. The time lapse between the shock arrival at Earth and Mars, for the events studied here, is shorter than expected from simulations, and the presence of high speed streams can enable an ICME to arrive almost simultaneously at Earth and Mars. This work will be applied to improve the input parameter methods for ENLIL.