We present nebular phase optical and near-infrared spectroscopy of the Type IIP supernova SN 2012aw combined with NLTE radiative transfer calculations applied to ejecta from stellar evolution/explosion models .
Our spectral synthesis models generally show good agreement with the ejecta from a M _ { ZAMS } = 15 M _ { \odot } progenitor star .
The emission lines of oxygen , sodium , and magnesium are all consistent with the nucleosynthesis in a progenitor in the 14 - 18 M _ { \odot } range .
We also demonstrate how the evolution of the oxygen cooling lines of [ O I ] \lambda 5577 , [ O I ] \lambda 6300 , and [ O I ] \lambda 6364 can be used to constrain the mass of oxygen in the non-molecularly cooled ashes to < 1 M _ { \odot } , independent of the mixing in the ejecta .
This constraint implies that any progenitor model of initial mass greater than 20 M _ { \odot } would be difficult to reconcile with the observed line strengths .
A stellar progenitor of around M _ { ZAMS } = 15 M _ { \odot } can consistently explain the directly measured luminosity of the progenitor star , the observed nebular spectra , and the inferred pre-supernova mass-loss rate .
We conclude that there is still no convincing example of a Type IIP supernova showing the nucleosynthesis products expected from a M _ { ZAMS } > 20 M _ { \odot } progenitor .