A portion of the Chandra Deep Field survey. Image: Credit: NASA/CXC/PSU/D.M.Alexander, F.E.Bauer, W.N.Brandt et al.

Large galaxies that should be in their prime have confounded expectations by experiencing less star formation relative to their mass than smaller galaxies that, according to hierarchical models of galaxy formation, should merely be the building blocks of those larger galaxies.

In a study presented today at the RAS National Astronomy Meeting at Queen's University Belfast, Bret Lehmer of the University of Durham presented findings of analysis of the Chandra Deep Field surveys (X-ray equivalents of the famous Hubble Deep Fields), looking for X-ray emitting galaxies. How bright galaxies are in X-rays is approximately proportional to the star-formation rate, with the majority of X-rays emitted by young, hot stars in binary systems that are accreting material onto a companion star. Because it is such a good measure of star formation, the analysis provides a clear picture of how star-formation changes with distance (and therefore time).

VIDEO Bret Lehmer of the University of Durham speaks with Astronomy Now Editor Keith Cooper about his research into star formation within galaxies.   PLAY

Lehmer, along with a team of international astronomers, took a sample of 2,568 galaxies between seven billion years ago and the present day. The galaxies covered all types, including regular spirals, starbursting galaxies and irregulars. Initially, only 225 were detected in X-rays, and of those 104 were normal galaxies like our Milky Way, and and 121 were active galaxies with energetic black holes within their cores, which could then be disregarded as they do no relate to star-formation rates. Then, taking the remaining 2,343 galaxies that were unaccounted for in X-rays, Lehmer and his team used a technique analogous to that used by amateur astrophotographers, where multiple images are stacked together so that their faint X-ray emission is gradually built up in increments and emerged from the background 'noise'.

What they found was eye-opening. Indeed, the X-ray emissions did change substantially over the last seven billion years, as the rate of star formation also changed (by a factor of up to ten for the most massive galaxies), but what was most surprising was that the most massive galaxies had the lowest X-ray emissions relative to their mass, at a time when they really should be showing a sharp rise in star formation and growing quite quickly through mergers with older, smaller galaxies. Instead, Lehmer and his team found that it was the smaller galaxies that instead displayed the most X-ray emission relative to their mass, and hence the greater percentage star-formation rate.

This has serious consequences for models of galactic evolution. The established model tens years ago described a Universe in where dwarf galaxies formed first, before merging with one another to build up larger galaxies with greater levels of star formation. That the smaller galaxies should have greater relative star-formation rates adds fuel to the fire that this theory of hierarchical formation is wrong.