Tests on artificial galaxies

An important issue in fitting profiles is the question how well the parametric fitting functions we use are able to reproduce the parameters values of the galaxy. To provide an answer the different fitting functions need to be tested on artificial galaxies for which we can set the parameter values as input. Empirically the fitting functions are quite capable to fit surface brightness profiles, but the question is if they really are able to converge to the input parameters.
We created a three dimensional model of an artificial galaxy with a vertical and radial exponential distribution. Radially we used a broken radial exponential structure (with typical values to the inner and outer scalelength ratio), to account for truncation. We made a model for a thin and a thick disk and added those together. We did line-of-sight integration over the radius to create a 2D artificial image, which will be filtered with the same methods as we use for the sample galaxies. We applied the same binning method as we did for our real data to create vertical profiles. The artifical galaxies we set up resemble our sample galaxies so that we are able to compare the fitting functions ability to reproduce the input parameters. The overall basic input parameters are $\mu_n$ = 19.24 mag arsec$^{-2}$, $z_n$ = 10, $h_n$ = 50 with the truncation, i.e. broken exponential, at 2.5$h_n$. The scaleheight of the thin disk was set sufficiently high so it would not be easily dominated by the thick disk component. The remaining parameters were varied to describe galaxies with two vertical disk components.