Understanding the evolution of galaxies strongly depends on our interpretation of their spectra. In the optical, BPT diagrams have been the main way to distinguish whether the principal excitation mechanism of a galaxy is dominated by star formation (SF) or by an active galactic nucleus (AGN). Although different classification methods exist, they are based on either hard-to-detect or high-energy emission lines. To date, the Balmer lines remain the most consistent way to classify galaxies, but at intermediate redshifts (1. 5 < z < 2. 5), galaxies are hard to parse in the BPT diagrams (and siblings) because the crucial halpha emission line is out of the range of ground-based optical spectrographs. In this work, we re-explore a diagram, which we call the OB-I diagram, that compares the equivalent width of hbeta and the emission line flux ratio of oiii/ and we breathe new life into it as it has the potential to be used for the classification of galaxies at these intermediate redshifts and to `clear the fog' that permeates galaxy classification in the optical rest frame. We used data from SDSS, LEGA-C, VANDELS, JADES, 3D-HST, and MOSDEF to explore galaxy classifications in the OB-I diagram across a wide range of redshifts (0 < z < 2. 7). Our results show that, at z < 0. 4, the OB-I diagram clearly separates galaxies between two distinct types: one dominated by an AGN and a second made up of a mixed population of SF galaxies and AGN activity. Comparison with the BPT diagrams and theoretical models shows that this mixed population can be partially separated from a pure SF population with a simple semi-empirical fit. At higher redshifts, we find that half of AGNs identified by other classification schemes are correctly recovered by the OB-I diagram, potentially making this diagram resistant to the cosmic shift that plagues most optical classification schemes, but more research is needed to understand this phenomenon. We find the OB-I diagram, which only requires two emission lines to be implemented, to be a useful tool at separating galaxies that possess a dominating AGN component in their emission from others. This applies not only to the Local Universe, but also seemingly at redshifts near the Cosmic Noon (z ∼ 2), without any need for significant adjustments in our empirical fit.
Santos et al. (Tue,) studied this question.