2025MNRAS.539.1723H

Supernova-driven galactic outflows play a vital but still poorly understood role in galactic chemical evolution, and one of the largest uncertainties about them is the extent to which they consist of supernova ejecta that are unmixed, or only poorly mixed, with the remainder of the interstellar medium (ISM). Simulations of wind launching make a range of predictions about the extent of mixing between the wind and the ISM, but thus far these have proven challenging to test observationally. In this study, we post-process high-resolution simulations of outflows from the QED simulation suite to generate synthetic X-ray spectra from galactic winds, which we then analyse using standard observational procedures, in order to search for detectable markers of wind mixing. Our synthetic observations reveal that partially mixed winds show significant and detectable metallicity gradients when viewed edge-on, with metallicity decreasing away from the central galactic disc. We explore how this signature results from imperfect mixing and the extent to which measurements of it can be used to diagnose the level of mixing in winds. We determine the signal-to-noise ratio (SNR) requirements for such measurements to be reliable, and provide a simple quantitative model that can be used to connect metallicity gradients to mixing between the hot ($T>10^{6}$ K) and cold ($T{<}10^{4}$ K) phases in observations that reach the required SNR, providing a framework to interpret current and future observations.