_version_ 1866916812222889984
author Cleri, Nikko J.
Olivier, Grace M.
Backhaus, Bren E.
Leja, Joel
Papovich, Casey
Trump, Jonathan R.
Haro, Pablo Arrabal
Buat, Veronique
Burgarella, Denis
Burnham, Emilie
Calabro, Antonello
Cohn, Jonathan H.
Cole, Justin W.
Davis, Kelcey
Dickinson, Mark
Finkelstein, Steven L.
Garner III, Ray
Hirschmann, Michaela
Hu, Weida
Hutchison, Taylor A.
Kocevski, Dale D.
Koekemoer, Anton M.
Larson, Rebecca L.
Lewis, Zach J.
Maseda, Michael V.
Seille, Lise-Marie
Simons, Raymond C.
author_facet Cleri, Nikko J.
Olivier, Grace M.
Backhaus, Bren E.
Leja, Joel
Papovich, Casey
Trump, Jonathan R.
Haro, Pablo Arrabal
Buat, Veronique
Burgarella, Denis
Burnham, Emilie
Calabro, Antonello
Cohn, Jonathan H.
Cole, Justin W.
Davis, Kelcey
Dickinson, Mark
Finkelstein, Steven L.
Garner III, Ray
Hirschmann, Michaela
Hu, Weida
Hutchison, Taylor A.
Kocevski, Dale D.
Koekemoer, Anton M.
Larson, Rebecca L.
Lewis, Zach J.
Maseda, Michael V.
Seille, Lise-Marie
Simons, Raymond C.
contents High-redshift observations from JWST indicate that optical strong line ratios do not carry the same constraining power as they do at low redshifts. Critically, this prevents a separation between stellar- and black hole-driven ionizing radiation, thereby obscuring both active galactic nuclei demographics and star formation rates. To investigate this, we compute a large suite of photoionization models from Cloudy powered by stellar populations and accreting black holes over a large grid of ages, metallicities, initial mass functions, binarity, ionization parameters, densities, and black hole masses. We use these models to test three rest-frame optical strong line ratio diagnostics which have been designed to separate ionizing sources at low redshifts: the [NII]-BPT, VO87, and OHNO diagrams. We show that the position of a model in these diagrams is strongly driven by the ionization parameter (log U) and the gas-phase metallicity, often more so than the ionizing spectrum itself; in particular, there is significant overlap between stellar population and accreting black hole models at high log U and low Z. We show that the OHNO diagram is especially susceptible to large contamination of the AGN region defined at z=1 for stellar models with high log U and low Z, consistent with many observed JWST spectra at high redshift. We show that the optical line ratio diagnostics are most sensitive to the shape of the <54 eV ionizing continuum, and that the derived ionizing sources for a given set of optical strong line ratios can be highly degenerate. Finally, we demonstrate that very high ionization (>54 eV) emission lines that trace ionizing sources harder than normal stellar populations help to break the degeneracies present when using the strong line diagnostics alone, even in gas conditions consistent with those at high redshifts.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21660
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Optical Strong Line Ratios Cannot Distinguish Between Stellar Populations and Accreting Black Holes at High Ionization Parameters and Low Metallicities
Cleri, Nikko J.
Olivier, Grace M.
Backhaus, Bren E.
Leja, Joel
Papovich, Casey
Trump, Jonathan R.
Haro, Pablo Arrabal
Buat, Veronique
Burgarella, Denis
Burnham, Emilie
Calabro, Antonello
Cohn, Jonathan H.
Cole, Justin W.
Davis, Kelcey
Dickinson, Mark
Finkelstein, Steven L.
Garner III, Ray
Hirschmann, Michaela
Hu, Weida
Hutchison, Taylor A.
Kocevski, Dale D.
Koekemoer, Anton M.
Larson, Rebecca L.
Lewis, Zach J.
Maseda, Michael V.
Seille, Lise-Marie
Simons, Raymond C.
Astrophysics of Galaxies
High-redshift observations from JWST indicate that optical strong line ratios do not carry the same constraining power as they do at low redshifts. Critically, this prevents a separation between stellar- and black hole-driven ionizing radiation, thereby obscuring both active galactic nuclei demographics and star formation rates. To investigate this, we compute a large suite of photoionization models from Cloudy powered by stellar populations and accreting black holes over a large grid of ages, metallicities, initial mass functions, binarity, ionization parameters, densities, and black hole masses. We use these models to test three rest-frame optical strong line ratio diagnostics which have been designed to separate ionizing sources at low redshifts: the [NII]-BPT, VO87, and OHNO diagrams. We show that the position of a model in these diagrams is strongly driven by the ionization parameter (log U) and the gas-phase metallicity, often more so than the ionizing spectrum itself; in particular, there is significant overlap between stellar population and accreting black hole models at high log U and low Z. We show that the OHNO diagram is especially susceptible to large contamination of the AGN region defined at z=1 for stellar models with high log U and low Z, consistent with many observed JWST spectra at high redshift. We show that the optical line ratio diagnostics are most sensitive to the shape of the <54 eV ionizing continuum, and that the derived ionizing sources for a given set of optical strong line ratios can be highly degenerate. Finally, we demonstrate that very high ionization (>54 eV) emission lines that trace ionizing sources harder than normal stellar populations help to break the degeneracies present when using the strong line diagnostics alone, even in gas conditions consistent with those at high redshifts.
title Optical Strong Line Ratios Cannot Distinguish Between Stellar Populations and Accreting Black Holes at High Ionization Parameters and Low Metallicities
topic Astrophysics of Galaxies
url https://arxiv.org/abs/2506.21660