The nature of giant clumps in distant galaxies probed by the anatomy of the cosmic snake

Abstract

Giant stellar clumps are ubiquitous in high-redshift galaxies. They are thought to play an important role in the build-up of galactic bulges and as diagnostics of star formation feedback in galactic discs. Hubble Space Telescope (HST) blank field imaging surveys have estimated that these clumps have masses up to 10$^{9.5}$ M$_{\odot}$ and linear sizes larger than ~1 kpc. Recently, gravitational lensing has also been used to get higher spatial resolution. However, both recent lensed observations and models suggest that the clumps properties may be overestimated by the limited resolution of standard imaging techniques. A definitive proof of this observational bias is nevertheless still missing. Here we investigate directly the effect of resolution on clump properties by analysing multiple gravitationally-lensed images of the same galaxy at different spatial resolutions, down to 30 pc. We show that the typical mass and size of giant clumps, generally observed at $\sim$1 kpc resolution in high-redshift galaxies, are systematically overestimated. The high spatial resolution data, only enabled by strong gravitational lensing using currently available facilities, support smaller scales of clump formation by fragmentation of the galactic gas disk via gravitational instabilities.