# Galaxy And Mass Assembly (GAMA): The environments of high- and low- excitation radio galaxies

Ching, JHY and Croom, SM and Sadler, EM and Robotham, ASG and Brough, S and Baldry, IK and Bland-Hawthorn, J and Colless, M and Driver, SP and Holwerda, BW and Hopkins, AM and Jarvis, MJ and Johnston, HM and Kelvin, LS and Liske, J and Loveday, J and Norberg, P and Pracy, MB and Steele, O and Thomas, D and Wang, L (2017) Galaxy And Mass Assembly (GAMA): The environments of high- and low- excitation radio galaxies. Monthly Notices of the Royal Astronomical Society. ISSN 0035-8711

We study the environments of low- and high- excitation radio galaxies (LERGs and HERGs respectively) in the redshift range $0.01 < z < 0.4$, using a sample of 399 radio galaxies and environmental measurements from the Galaxy And Mass Assembly (GAMA) survey. In our analysis we use the fifth nearest neighbour density ($\Sigma_{5}$) and the GAMA galaxy groups catalogue (G3Cv6) and construct control samples of galaxies matched in {\update stellar mass and colour} to the radio-detected sample. We find that LERGs and HERGs exist in different environments and that this difference is dependent on radio luminosity. High-luminosity LERGs ($L_{\rm NVSS} \gtrsim 10^{24}$ W Hz$^{-1}$) lie in much denser environments than a matched radio-quiet control sample (about three times as dense, as measured by $\Sigma_{5}$), and are more likely to be members of galaxy groups ($82^{+5}_{-7}$ percent of LERGs are in GAMA groups, compared to $58^{+3}_{-3}$ percent of the control sample). In contrast, the environments of the HERGs and lower luminosity LERGs are indistinguishable from that of a matched control sample. Our results imply that high-luminosity LERGs lie in more massive haloes than non-radio galaxies of similar stellar mass and colour, in agreement with earlier studies (Wake et al. 2008; Donoso et al. 2010). When we control for the preference of LERGs to be found in groups, both high- and low- luminosity LERGs are found in higher-mass haloes ($\sim 0.2$ dex; at least 97 percent significant) than the non-radio control sample.