Herschel observations of extended atomic gas in the core of the Perseus cluster

Abstract

We present Herschel observations of the core of the Perseus cluster of galaxies. Especially intriguing is the network of filaments that surround the brightest cluster galaxy, NGC 1275, previously imaged extensively in Hα and CO. In this work, we report detections of far-infrared (FIR) lines, in particular, [C ii] 158, [O i] 63, [N ii] 122, [O ib] 145 and [O iii] 88  μm, with Herschel. All lines are spatially extended, except [O iii], with the [C ii] line emission extending up to 25 kpc from the core. [C ii] emission is found to be co-spatial with Hα and CO. Furthermore, [C ii] shows a similar velocity distribution to CO, which has been shown in previous studies to display a close association with the Hα kinematics. The spatial and kinematical correlation among [C ii], Hα and CO gives us confidence to model the different components of the gas with a common heating model. With the help of FIR continuum Herschel measurements, together with a suite of coeval radio, sub-millimetre and IR data from other observatories, we performed a spectral energy distribution fitting of NGC 1275 using a model that contains contributions from dust emission as well as synchrotron active galactic nucleus emission. This has allowed us to accurately estimate the dust parameters. The data indicate a low dust emissivity index, β ≈ 1, a total dust mass close to 107 M⊙, a cold dust component with temperature 38 ± 2 K and a warm dust component with temperature 116 ± 9 K. The FIR-derived star formation rate is 24 ± 1 M⊙ yr−1, which is in agreement with the far-ultraviolet-derived star formation rate in the core, determined after applying corrections for both Galactic and internal reddening. The total IR luminosity in the range 8–1000  μm is inferred to be 1.5 × 1011 L⊙, making NGC 1275 a luminous IR galaxy. We investigated in detail the source of the Herschel FIR and Hα emissions emerging from a core region 4 kpc in radius. Based on simulations conducted using the radiative transfer code, cloudy, a heating model comprising old and young stellar populations is sufficient to explain these observations. The optical line ratios indicate that there may be a need for a second heating component. However, stellar photoionization seems to be the dominant mechanism. We have also detected [C ii] in three well-studied regions of the filaments. Herschel, with its superior sensitivity to FIR emission, can detect far colder atomic gas than previous studies. We find an [O i]/[C ii] ratio about 1 dex smaller than predicted by the otherwise functional Ferland (2009) model. That study considered optically thin emission from a small cell of gas and by design did not consider the effects of reasonable column densities. The line ratio suggests that the lines are optically thick, as is typical of galactic photodissociation regions, and implies that there is a large reservoir of cold atomic gas. This was not included in previous inventories of the filament mass and may represent a significant component.