Using 21 cm absorption surveys to measure the average H i spin temperature in distant galaxies

• Published in: Monthly notices of the Royal Astronomical Society Vol. 462; no. 2; pp. 1341 - 1350
• Main Authors: Allison, J. R., Zwaan, M. A., Duchesne, S. W., Curran, S. J.
• Format: Journal Article
• Language: English
• Published: London Oxford University Press 21-10-2016
• Subjects: Absorption; Arrays; Astronomy; Cold spinning; Cosmology; Frequency distribution; Galaxies; Radio sources (astronomy); Red shift; Simulation; Spin temperature; Star & galaxy formation; Statistical methods; Australia; radio lines: galaxies; methods: statistical; galaxies: evolution; quasars: absorption lines; galaxies: ISM; galaxies: high-redshift
• ISSN: 0035-8711; 1365-2966
• DOI: 10.1093/mnras/stw1722

We present a statistical method for measuring the average H i spin temperature in distant galaxies using the expected detection yields from future wide-field 21 cm absorption surveys. As a demonstrative case study, we consider an all-southern-sky simulated survey of 2-h per pointing with the Australian Square Kilometre Array Pathfinder for intervening H i absorbers at intermediate cosmological redshifts between z = 0.4 and 1. For example, if such a survey yielded 1000 absorbers, we would infer a harmonic-mean spin temperature of
$\overline{T}_\mathrm{spin} \sim 100$
 K for the population of damped Lyman α absorbers (DLAs) at these redshifts, indicating that more than 50 per cent of the neutral gas in these systems is in a cold neutral medium (CNM). Conversely, a lower yield of only 100 detections would imply
$\overline{T}_\mathrm{spin} \sim 1000$
 K and a CNM fraction less than 10 per cent. We propose that this method can be used to provide independent verification of the spin temperature evolution reported in recent 21 cm surveys of known DLAs at high redshift and for measuring the spin temperature at intermediate redshifts below z ≈ 1.7, where the Lyman α line is inaccessible using ground-based observatories. Increasingly more sensitive and larger surveys with the Square Kilometre Array should provide stronger statistical constraints on the average spin temperature. However, these will ultimately be limited by the accuracy to which we can determine the H i column density frequency distribution, the covering factor and the redshift distribution of the background radio source population.