Abstract. Our project, 'MegaMorph', is developing a next-generation tool for decomposing galaxies, in terms of both their structures and stellar populations. By combining data from UV to NIR wavelengths, accounting for morphological peculiarities using non-parametric components, and utilising efficient likelihood sampling methods, we are working to significantly improve the robustness and accuracy of galaxy decomposition. Applying these new techniques to modern large surveys will provide us with a deeper understanding of galaxies.Keywords. galaxies: fundamental parameters, galaxies: structure, galaxies: photometry, methods: data analysis, techniques: image processing
Galaxy componentsGalaxies comprise a complex mix of stars, gas, dust and dark matter, with their proportions, distributions and kinematics resulting from a wide variety of physical processes. Even considering just the stars, galaxies typically possess distinct stellar populations, which formed at different times, under varied conditions. To gain more than a superficial understanding of the galaxy population, and the processes that shape it, we need to disentangle these physical components within individual galaxies.This conference primarily focused on separating and characterising galaxy components through their spectral energy distributions. These SEDs are usually, with a few notable exceptions, integrated over the spatial extent of each galaxy. For separating components which dominate at different wavelength ranges, e.g., stars and dust, this approach works well. However, when dividing components with only subtle differences in their SEDs, e.g., multiple stellar populations, any additional discriminatory information is valuable.The various stellar populations within a galaxy often have contrasting spatial distributions. In our simplest models, a galaxy grows through two principal mechanisms: the gradual accretion of gas, which cools and forms a thin disk of stars; and the merging of existing stellar systems, which produces a spheroid. These two structural components thus form through disparate mechanisms, contain different stellar populations, and represent disjoint periods in a galaxy's history (e.g., Cole et al. 2000;Cook et al. 2009;Benson 2010). Using information regarding both the spatial distributions and SEDs of these components will clearly enable a more robust separation of their properties, when compared with trying to use just one of these pieces of information alone. The dichotomy between disk and spheroid stellar components (e.g., Allen et al. 2006;Benson & Devereux 2010) is at the heart of our understanding of galaxy development: measuring their individual properties will provide a deeper understanding of the galaxy population and more effective comparison with theoretical models.