We present a Monte-Carlo package for simulation of high-redshift supernova data, SNOC. Optical and near-infrared photons from supernovae are ray-traced over cosmological distances from the simulated host galaxy to the observer at Earth. The distances to the sources are calculated from user provided cosmological parameters in a Friedmann-Lemaitre universe, allowing for arbitrary forms of ``dark energy''. The code takes into account gravitational interactions (lensing) and extinction by dust, both in the host galaxy and in the line-of-sight. The user can also choose to include exotic effects like a hypothetical attenuation due to photon-axion oscillations. SNOC is primarily useful for estimations of cosmological parameter uncertainties from studies of apparent brightness of Type Ia supernovae vs redshift, with special emphasis on potential systematic effects. It can also be used to compute standard cosmological quantities like luminosity distance, lookback time and age of the universe in any Friedmann-Lemaitre model with or without quintessence.

The use of Type Ia supernovae as calibrated standard candles is one of the most powerful tools to study the expansion history of the universe and thereby its energy components. While the analysis of some ~50 supernovae at redshifts around z~0.5 have provided strong evidence for an energy component with negative pressure, ``dark energy'', more data is needed to enable an accurate estimate of the amount and nature of this energy. This might be accomplished by a dedicated space telescope, the SuperNova / Acceleration Probe (2000; SNAP), which aims at collecting a large number of supernovae with z<2. In this paper we assess the ability of the SNAP mission to determine various properties of the ``dark energy.'' To exemplify, we expect SNAP, if operated for three years to study Type Ia supernovae, to be able to determine the parameters in a linear equation of state w(z)=w0 + w1 z to within a statistical uncertainty of +-0.04 for w0 and +0.15,-0.17 for w1 assuming that the universe is known to be flat and an independent high precision (sigma_{Omega_m}=0.015) measurement of the mass density Omega_m, is used to constrain the fit. An additional improvement can be obtained if a large number of low-z, as well as high-z, supernovae are included in the sample.

Recently, Holz and Wald (1998) have presented a method for determining gravitational lensing effects in inhomogeneous universes. Their use of realistic galaxy models has been limited to the singular, truncated isothermal sphere with a fixed mass. In this paper, their method is generalized to allow for matter distributions more accurately describing the actual properties of galaxies, as derived from observations and N-body simulations. This includes the density profile proposed by Navarro, Frenk and White, as well as a distribution of galaxy masses. As an example of the possible applications of the method, we consider lensing effects on supernova luminosity distributions. We find that results for different mass distributions of smooth dark matter halos are very similar, making lensing effects predictable for a broad range of halo profiles. We also note, in agreement with other investigations, that one should be able to discriminate smooth halos from a dominant component of dark matter in compact objects. For instance, a sample of 100 supernovae at redshift z=1 can, with 99 % certainty, discriminate the case where all matter is in compact objects from the case where matter is in smooth halos.

Recently, Holz & Wald have presented a new method for determining gravitational lensing effects on, e.g., supernova luminosity versus redshift measurements in inhomogeneous universes. In this paper, their method is generalized in several ways: First, the matter content is allowed to consist of several different types of fluids, possibly with non-vanishing pressure. Second, besides lensing by simple point masses and singular isothermal spheres, the more realistic halo dark matter distribution proposed by Navarro, Frenk & White (NFW), based on N-body simulation results, is treated. We discuss various aspects of the accuracy of the method, such as luminosity corrections, and statistics, for multiple images. We find in agreement with other recent work that a large sample of supernovae at large redshift could be used to extract gross features of the mass distribution of the lensing dark matter halos, such as the existence of a large number of point-like objects. The results for the isothermal sphere and the NFW model are, however, very similar if normalized to the observed luminosity distribution of galaxies. We give convenient analytical fitting formulas for our computed lensing probabilites as a function of magnification, for several redshifts.

Spatially homogeneous cosmological models with a positive cosmological constant are investigated, using dynamical systems methods. We focus on the future evolution of these models. In particular, we address the question whether there are models within this class that are de Sitter-like in the future, but are tilted.