Emission-line activity in type 2 quasars from the Sloan Digital Sky Survey

We have compared the optical emission-line ratios of type 2 quasars from Zakamska et al. with standard active galactic nucleus (AGN) photoionization model predictions, type 2 Seyfert galaxies, H galaxies and narrow-line Fanaro-Riley type II radio galaxies. Moderate to high-ionization narrow-line radio galaxies and type 2 Seyfert galaxies are indistinguishable from type 2 quasars based on their optical line ratios. The standard AGN photoionization models, widely discussed for other type 2 AGN, can reproduce successfully the loci and trends of type 2 quasars in some of the main diagnostic diagrams. These models are not exempt of problems, and the discrepancies with the data are the same encountered for other type 2 AGN. As for these, realistic models must take into account a range of cloud properties, as widely demonstrated in the literature. The Zakamska et al. sample is strongly biased towards objects with high line luminosities (L[O-III] > 10(42) erg s(-1)). We have found that stellar photoionization is obvious in a fraction of objects (3 out of 50) which are characterized by low [O-III] luminosities compared with most type 2 quasars in the sample. We suggest that if the sample were expanded towards lower line luminosities (similar to 10(40-42) erg s(-1)) stellar photoionization might be evident much more frequently. We explore an alternative scenario to pure AGN photoionization in which a varying contribution of stellar ionized gas contributes to the line fluxes. Although the models presented here are rather simplistic and not strong quantitative results can be extracted regarding the relative contribution of stellar versus AGN photoionization, our results suggest that adding a varying contribution of stellar photoionized gas works in the right direction to solve most of the problems affecting the standard AGN photoionization models. The 'temperature problem' on the other hand remains.