The nature of the progenitor of the Type II-P supernova 1999em

The masses and the evolutionary states of the progenitors of core-collapse supernovae (SNe) are not well constrained by direct observations. Stellar evolution theory generally predicts that massive stars with initial masses less than about 30 M. should undergo core collapse when they are cool M-type supergiants. However, the only two detections of an SN progenitor before explosion are SN 1987A and SN 1993J, and neither of these was an M-type supergiant. Recently, we have set an upper mass limit to the progenitor of Type II-P SN 1999gi of 9(-2)(+3) M. from preexplosion Hubble Space Telescope images. In this paper we present high-quality ground-based VRI images of the site of the Type II-P SN 1999em (in NGC 1637) taken before explosion, which were extracted from the Canada-France-Hawaii Telescope archive. We determine a precise position of the SN on these images to an accuracy of The host 0."17. galaxy is close enough (7.5 +/- 0.5 Mpc) that the bright supergiants are resolved as individual objects; however, we show that there is no detection of an object at the SN position before explosion that could be interpreted as the progenitor star. By determining the sensitivity limits of the VRI data and assuming that the reddening toward the progenitor is similar to that toward the SN itself, we derive bolometric luminosity limits for the progenitor. Comparing these to standard stellar evolutionary tracks that trace evolution up to the point of core carbon ignition, we initially derive an upper mass limit of approximately 12 M.. However, we present evolutionary calculations that follow 7-12 M. stars throughout their C-burning lifetime and show that we can restrict the mass of the progenitor even further. Our calculations indicate that progenitors initially of 8-10 M. undergoing expected mass loss can also be excluded because a second dredge-up sends them to somewhat higher luminosities than a star of initially 12 M.. These results limit the progenitor's initial main-sequence mass to a very narrow range of 12(-1)(+1) M.. We discuss the similarities between the Type II-P SNe 1999em and 1999gi and their progenitor mass limits and also consider all the direct evidence currently available on progenitors of core-collapse SNe. We suggest that SNe Type II-P originate only in intermediate mass stars of 8-12 M. which are in the red supergiant region, and that higher mass stars produce the other Type II subtypes. Finally, we present a discussion on the future possibilities of determining masses or mass limits and the evolutionary status of core-collapse events on a statistically larger sample, using virtual observatory initiatives such as ASTROVIRTEL.