Testing Diagnostics of Nuclear Activity and Star Formation in Galaxies at z>1

TL;DR

This study evaluates the effectiveness of various AGN and star formation diagnostics at redshift ~1.5 using new infrared spectra from Keck/MOSFIRE, comparing traditional and alternative methods with X-ray data.

Contribution

It demonstrates the reliability of classic emission-line diagnostics at z>1 and highlights the need to recalibrate color-excitation methods for high-redshift galaxy classification.

Findings

High [OIII]/Hb ratio alone is insufficient for AGN identification at z>1.

Classic diagnostics remain consistent with X-ray classifications for detected galaxies.

Color-excitation method requires recalibration to be effective at high redshift.

Abstract

We present some of the first science data with the new Keck/MOSFIRE instrument to test the effectiveness of different AGN/SF diagnostics at z~1.5. MOSFIRE spectra were obtained in three H-band multi-slit masks in the GOODS-S field, resulting in two hour exposures of 36 emission-line galaxies. We compare X-ray data with the traditional emission-line ratio diagnostics and the alternative mass-excitation and color-excitation diagrams, combining new MOSFIRE infrared data with previous HST/WFC3 infrared spectra (from the 3D-HST survey) and multiwavelength photometry. We demonstrate that a high [OIII]/Hb ratio is insufficient as an AGN indicator at z>1. For the four X-ray detected galaxies, the classic diagnostics ([OIII]/Hb vs. [NII]/Ha and [SII]/Ha) remain consistent with X-ray AGN/SF classification. The X-ray data also suggest that "composite" galaxies (with intermediate AGN/SF classification) host bona-fide AGNs. Nearly 2/3 of the z~1.5 emission-line galaxies have nuclear activity detected by either X-rays or the classic diagnostics. Compared to the X-ray and line ratio classifications, the mass-excitation method remains effective at z>1, but we show that the color-excitation method requires a new calibration to successfully identify AGNs at these redshifts.