EXPLOSION MODELS FOR TYPE IA SUPERNOVAE - A COMPARISON WITH OBSERVED LIGHT CURVES, DISTANCES, H-0, AND Q(0)

Theoretical monochromatic light curves and photospheric expansion velo cities are compared with observations of 27 Type Ia supernovae (SN Ia' s). A set of 37 models has been considered which encompasses all curre ntly discussed explosion scenarios for Type Ia supernovae including de flagrations, detonations, delayed detonations, pulsating delayed deton ations and tamped detonations of Chandrasekhar mass, and helium detona tions of low-mass white dwarfs. The explosions are calculated using on e-dimensional Lagrangian hydro and radiation-hydro codes with incorpor ated nuclear networks. Subsequently, light curves are constructed usin g our light-curve (LC) scheme which includes an implicit radiation tra nsport, expansion opacities, a Monte Carlo gamma-ray transport, and mo lecular and dust formation. For some supernovae, results of detailed n on-LTE calculations have been considered. Observational properties of our series of models are discussed, in particular, the relation betwee n the absolute brightness, postmaximum decline rates, the colors at se veral moments of time, etc. All models with a Ni-56 production larger than approximate to 0.4 M. produce light curves of similar brightness. The influence of the cosmological redshift on the light curves and on the correction for interstellar reddening is discussed. Based on data rectification of the standard deviation, a quantitative procedure to fit the observations has been used to determine the free parameters, i .e., the distance, the reddening, and the time of the explosion. Fast- rising light curves (e.g., SN 1981B and SN 1994D) can be reproduced by delayed detonation models or deflagration models similar to W7. Slowl y rising (t(max) greater than or equal to 16 days) light curves (e.g., SN 1984A and SN 1990N) cannot be reproduced by standard detonation, d eflagration, or delayed detonation models. To obtain an acceptable agr eement with observations, models are required in which the C/O white d warf is surrounded by an unburned extended envelope of typically 0.2-0 .4 M. which may either be preexisting or produced during the explosion . Our interpretation of the light curves is also supported by the phot ospheric expansion velocities. Mainly due to the fast increase of the gamma radiation produced by the outer Ni-56, the postmaximum decline o f helium detonations tends to be faster compared to observations of no rmal bright SN Ia's. Strongly subluminous SN Ia's can be understood in the framework of pulsating delayed detonations, both from the absolut e brightness and the colors. Alternatively, subluminosity can be produ ced within the scenario of helium detonations in low-mass white dwarfs of about 0.6-0.8 M. if the explosion occurs when rather little helium has been accreted. However, even subluminous helium detonation models are very blue at maximum light owing to heating in the outer layers, and brighter models show a fast postmaximum decline, in contradiction to the observations. We find evidence for a correlation between the ty pe of host galaxy and the explosion mechanism. In spiral galaxies, abo ut the same amount of prompt explosions (delayed detonations and W7) a nd pulsating delayed detonations seems to occur. In contrast, in ellip ticals, the latter type is strongly favored. This difference may provi de a hint about the stellar evolution of the progenitors. Based on a c omparison of theoretical light curves and observational data, the dist ances of the parent galaxies are determined independently from seconda ry distance indicators. A comparison with theoretical models allows fo r a consistent determination of the interstellar reddening and the cos mological redshift. For the example of SN 1988U, we show the need for a simultaneous use of both spectral and light curve data if the data s et is incomplete. Based on the models, SN Ia's allow for a measurement of the value of the Hubble constant H-0. H-0 is found to be 67 +/- 9 km s(-1) Mpc(-1) within a 95% probability for distances up to 1.3 Gpc. -1 SN 1988U at 1.3 Gpc is consistent with a deceleration parameter q( 0) of 0.7 +/- 0.5 (1 sigma).