| Upcoming blind HI surveys with e.g. APERTIF or ASKAP will provide data for tens of thousands of galaxies that can be used for a detailed study of galaxy evolution. Due to the large number of expected sources and the limited resolution of the majority of objects, detailed modelling is not feasible for most detections. The use of non-parametric methods for the classification of objects is a promising alternative because they are fast and objective. Unfortunately they suffer from effects caused by low signal-to-noise, limited resolution as well as the intrinsic orientation of the studied objects. We investigated the limitations of existing non-parametric methods and present a possible method for the correction of the effects of signal-to-noise.
The figure clearly shows for two sets of model galaxies how well the signal-to-noise correction works on the Asymmetry parameter (lower panels, to be compared to the upper panels with uncorrected asymmetries) even for many of the the low signal-to-noise (blue) objects.
N. Giese, T. van der Hulst, P. Serra, T. Oosterloo: "Non-parametric estimation of morphological lopsidedness", accepted for publication in MNRAS
|N. Giese, T. van der Hulst, P. Serra, T. Oosterloo, Image of WSRT: ASTRON