2005A&A...434...77L

We present a quantitative method to classify galaxies, based on multi-wavelength data and constructed from the properties of nearby galaxies. Our objective is to define an classification method that can be used for low and high redshift objects. We estimate the concentration of light (C) at the galaxy center and the 180° rotational asymmetry (A), computed at several wavelengths, from ultraviolet (UV) to I-band. The variation of the indices of concentration and asymmetry with the wavelength reflects the proportion and the distribution of young and old stellar populations in galaxies. In general C is found to decrease, and A to increase from optical to UV: the patchy appearance of a galaxy in the UV with no bulge is often very different from its counterpart at optical wavelengths, with a prominent bulge and a more regular disk. We quantify the variation of C and A with wavelength. In this way we are able to distinguish five types of galaxies that we call spectro-morphological types: compact, ringed, spiral, irregular and central-starburst galaxies, which can be differentiated by the distribution of their stellar populations. We discuss in detail the morphology of the galaxies of the sample, and describe the morphological characteristics of each spectro-morphological type. We apply spectro-morphology to three objects at a redshift z∼1 in the Hubble Deep Field North, which gives encouraging results for applications to large samples of high-redshift galaxies. This method of morphological classification could be used to study the evolution of the morphology with redshift and is expected to put observational constraints on scenarios of galaxy evolution.