We present results from an early Eocene (ca. 50–56 Ma ago) regional modeling sensitivity experiment that examines the role of atmospheric CO2 in determining the regional climate of western North America. This is the first paleoclimate modeling study to investigate the possible role of increased CO2 in influencing the early Eocene climate on a regional scale. We take a regional modeling approach with the goal of using higher spatial resolution to elucidate the role of specific climate forcing mechanisms (here, CO2) upon a region with relatively dense paleoclimate proxy data coverage. The spatial resolution of global climate models does not permit the close comparison of model results to proxy climate data in a way that helps to distinguish between regional and global climate forcings, which is a goal of this study. While our results suggest that CO2 was most likely at least as high as 2240 ppm, this high concentration does not yield a regional climate that matches regional proxy data in all aspects. Therefore, in combination with high atmospheric CO2, other forcing factors must have played significant roles in defining the nature of early Eocene climate.