Next Contents Previous


Technically, comparison between observations and simulations is relatively straightforward. From the coordinates of the particles in the luminous components, and after choosing the viewing angles and taking into account the observational conditions, it is possible to obtain an image that can be output in the standard format used by observers, namely FITS (Flexible Image Transport System). This image can then be analysed as are observations, using standard packages, such as IRAF (Image Reduction and Analysis Facility). Similarly one can create data cubes, containing velocity information, which again will be analysed with the same software packages as observations. Taking into account the limitations of the instruments and more generally those due to observational conditions is an important feature here, as is the fact that it is the simulation data that must be transformed into observations and not the other way round.

There is, nevertheless, one subtle point concerning a limitation of dynamical simulations that should be kept in mind. It concerns the simulation time to be chosen for the comparison. As already mentioned, in dynamic simulations the disk is assumed to be in place and in equilibrium before the bar starts forming. On the contrary, in cosmological simulations the bar should start forming as soon as the relative disk mass is sufficiently high to allow the bar instability to proceed. One must add to this the uncertainty about when disks can be considered as being in place. All this taken together makes it very difficult to pinpoint the simulation time to be used for the comparisons. The best is to try a range of times and then describe how the fit evolves with time.