The impact of Lyman-α radiative transfer on large-scale clustering in the Illustris simulation
Christoph Behrens, Chris Byrohl, Shun Saito, Jens Niemeyer
Astronomy & Astrophysics , 614 , A31 · June 2018
Abstract
Context: Lyman-α emitters (LAEs) are a promising probe of large-scale structure at high redshift, z ≳ 2. In particular, the Hobby-Eberly Telescope Dark Energy Experiment aims to observe LAEs at 1.9 < z < 3.5 to measure the baryon acoustic oscillation scale and redshift-space distortion. However, the complicated radiative transfer (RT) of the resonant Lyman-α emission line may generate an anisotropic selection bias in LAE clustering on large scales. We aim to quantify the impact of Lyman-α RT on large-scale galaxy clustering by studying correlations between the large-scale environment and the ratio of apparent to intrinsic Lyman-α luminosity. We apply our Lyman-α RT code by post-processing the full Illustris simulations. We find little correlation between large-scale environment and the observed fraction induced by RT, and hence a smaller anisotropic selection bias than previously claimed. We argue that the anisotropy was overestimated in previous work because of insufficient spatial resolution; resolving the high-density region down to interstellar-medium scales is important. We also find that the correlation can be further enhanced by assumptions in modeling intrinsic Lyman-α emission.
Motivation
Commonly galaxy surveys account for distortions of the clustering signal arising from the most common redshift space distortions: the Fingers-of-God effect and the Kaiser effect.
For the Lyman- emission line (Zheng et al. 2011) found a further distortion effect arising from radiative transfer that occurs in real space (and is therefore a selection effect). In numerical simulations this effect was found to show up in the clustering signal by elongation along the line of sight and a decreased bias. This effect can be attributed to correlations of the escape fraction of photons through the IGM with the density and velocity gradient. Given these correlations, the distortions effect are not just limited to small scales, but need to be accounted for in studies aiming to constrain the cosmological model from the large scale modes in Lyman- galaxy surveys and intensity maps, such as in the HETDEX survey.
Summary
In this paper, we reevaluated the nature of this effect and find that it is strongly tied to the spectral shape arising on CGM scales. As the prior publication does not sufficiently resolve this scale, our result vastly changes as we increase the resolution by two orders of magnitude based on the Illustris simulations. The distortion (selection) effect from radiative transfer disappears as we increase the resolution. A representative plot of this result can be found above.
While we demonstrated in this paper that there is no radiative transfer distortion effect in our simulations, this does not exclude the possible existence of such effect in reality: The effect strongly depends on the spectrum on leaving the CGM. Dust and clumpiness might strongly effect this spectrum and thus the strength of a possible distortion effect. Future improvements in understanding in the multi-scale modeling (ISM/CGM/IGM) of Lyman- emitter observations will help settling this question.