Galaxy pairs in the Sloan Digital Sky Survey - IV. Interactions trigger active galactic nuclei

Galaxy-galaxy interactions are predicted to cause gas inflows leading to enhanced nuclear star formation. This prediction is borne out observationally, and is also supported by the gas-phase metallicity dilution in the inner regions of galaxies in close pairs. In this paper we test the further prediction that the gas inflows lead to enhanced accretion on to the central supermassive black hole, triggering activity in the nucleus. Based on a sample of 11060 Sloan Digital Sky Survey galaxies with a close companion (r_p < 80h^-1₇₀ kpc, ΔV < 200kms^-1), we classify active galactic nuclei (AGN) based either on emission line ratios or on spectral classification as a quasar. The AGN fraction in the close pairs sample is compared to a control sample of 110600 mass- and redshift-matched control galaxies with no nearby companion. We find a clear increase in the AGN fraction in close pairs of galaxies with projected separations < 40 h^-1₇₀ kpc by up to a factor of 2.5 relative to the control sample [although the enhancement depends on the chosen signal-to-noise ratio (S/N) cut of the sample]. The increase in AGN fraction is strongest in equal-mass galaxy pairings, and weakest in the lower mass component of an unequal-mass pairing. The increased AGN fraction at small separations is accompanied by an enhancement in the number of 'composite' galaxies whose spectra are the result of photoionization by both AGN and stars. Our results indicate that AGN activity occurs (at least in some cases) well before final coalescence and concurrently with ongoing star formation. Finally, we find a marked increase at small projected separations of the fraction of pairs in which both galaxies harbour AGN. We demonstrate that the fraction of double AGN exceeds the expected random fraction, indicating that some pairs undergo correlated nuclear activity. We discuss some of the factors that have led to conflicting results in previous studies of AGN in close pairs. Taken together with complementary studies, we favour an interpretation where interactions trigger AGN, but are not the only cause of nuclear activity.