Claims of an accelerated cosmic expansion and thus a new constituent of the universe are entirely based on supernova observations. The geometric measure of the universe has been achieved with SNe Ia as distance indicators out to a redshift of 1. This result has important consequences for all of physics, because the acceleration would be caused by either a cosmological constant or a yet unknown new form of `dark energy'. However, the luminosity of nearby supernovae varies by up to a factor of ten, and it is only by means of empirical relations between luminosity and distance-independent properties, like the shape of the light curves, that supernovae can be turned into good distance indicators even for high redshifts.
Currently, our understanding of the supernova physics is the major systematic uncertainty of this result. The distant supernovae exploded at a time when our solar system was just forming. There is no guarantee that these distant explosions are the same as the nearby ones on which the empirical relations are based. Only once we understand the physics of the explosions will we be able to assess whether they can be used as distance indicators reliably, and whether we have to search for new physics beyond the standard models of particle physics and cosmology.
The present proposal aims at a major break-through in our understanding of Type Ia supernovae in the coming years. This will be achieved by a combined action of all major groups working in this field in Europe, for the first time being in a position to obtain the necessary increase of homogeneous sets of lightcurves and spectra for a large sample of Type Ia supernovae by both, systematic and targeted observations and by increasing the quality and predictive power of theoretical models at the same time.