Optical and infrared observations of the Type IIPSN 2002hh from days 3 to 397

We present optical and infrared (IR) observations of the Type II SN 2002hh from 3 to 397 d after explosion. The optical spectroscopic (4-397 d) and photometric (3-278 d) data are complemented by spectroscopic (137-381 d) and photometric (137-314 d) data acquired at IR wavelengths. This is the first time L-band spectra have ever been successfully obtained for a supernova (SN) at a distance beyond the Local Group. The VRI light curves in the first 40 d reveal SN 2002hh to be an SN IIP (plateau) - the most common of all core-collapse SNe. SN 2002hh is one of the most highly extinguished SNe ever investigated. To provide a match between its early-time spectrum and a coeval spectrum of the Type IIP SN 1999em, as well as maintaining consistency with K I interstellar absorption, we invoke a two-component extinction model. One component is due to the combined effect of the interstellar medium (ISM) of our Milky Way Galaxy and the SN host galaxy, while the other component is due to a 'dust pocket' where the grains have a mean size smaller than in the ISM. The early-time optical light curves of SNe 1999em and 2002hh are generally well matched, as are the radioactive tails of these two SNe and SN 1987A. The late-time similarity of the SN 2002hh optical light curves to those of SN 1987A, together with measurements of the optical/IR luminosity and [Fe II] 1.257 mu m emission indicate that 0.07 +/- 0.02 M-circle dot of Ni-56 was ejected by SN 2002hh. However, during the nebular phase the HKL' luminosities of SN 2002hh exhibit a growing excess with respect to those of SN 1987A. We attribute much of this excess to an IR-echo from a pre-existing, dusty circumstellar medium. Based on an IR-echo interpretation of the near-IR (NIR) excess, we deduce that the progenitor of SN 2002hh underwent recent mass-loss of similar to 0.3 M-circle dot. A detailed comparison of the late-time optical and NIR spectra of SNe 1987A and 2002hh is presented. While the overall impression is one of similarity between the spectra of the two events, there are notable differences. The Mg I 1.503 mu m luminosity of SN 2002hh is a factor of 2.5 greater than in SN 1987A at similar epochs, yet coeval silicon and calcium lines in SN 2002hh are fainter. Interpreting these differences as being due to abundance variations, the overall abundance trend between SN 1987A and 2002hh is not consistent with explosion model predictions. It appears that during the burning to intermediate-mass elements, the nucleosynthesis did not progress as far as might have been expected given the mass of iron ejected. Evidence for mixing in the ejecta is presented. Pronounced blueshifts seen in the more isolated lines are attributed to asymmetry in the ejecta. However, during the time-span of these observations (similar to 1-yr post-explosion) we find no evidence of dust condensation in the ejecta such as might have been revealed by an increasing blueshift and/or attenuation of the red wings of the emission lines. Nevertheless, the clear detection of first overtone COemission by 200 d and the reddening trend in (K-L')(0) suggest that dust formation in the ejecta may occur at later epochs. From the[OI] lambda lambda 6300, 6364 angstrom doublet luminosity we infer a 16-18 M-circle dot main-sequence progenitor star. The progenitor of SN 2002hh was probably a red supergiant with a substantial, dusty wind.