Zeta Oph and the weak-wind problem

Mass-loss rate, $\dot{M}$, is one of the key parameters affecting evolution and observational manifestations of massive stars, and their impact on the ambient medium. Despite its importance, there is a factor of ~100 discrepancy between empirical and theoretical $\dot{M}$ of late-type O dwarfs, the so-called weak-wind problem. In this Letter, we propose a simple novel method to constrain $\dot{M}$ of runaway massive stars through observation of their bow shocks and Stromgren spheres, which might be of decisive importance for resolving the weak-wind problem. Using this method, we found that $\dot{M}$ of the well-known runaway O9.5 V star zeta Oph is more than an order of magnitude higher than that derived from ultraviolet (UV) line-fitting (Marcolino et al. 2009) and is by a factor of 6 to 7 lower than those based on the theoretical recipe by Vink et al. (2000) and the Halpha line (Mokiem et al. 2005). The discrepancy between $\dot{M}$ derived by our method and that based on UV lines would be even more severe if the stellar wind is clumpy. At the same time, our estimate of $\dot{M}$ agrees with that predicted by the moving reversing layer theory by Lucy (2010).