The Number Density Evolution of Extreme Emission Line Galaxies in 3D-HST: Results from a Novel Automated Line Search Technique for Slitless Spectroscopy

TL;DR

This paper introduces a new Bayesian method for detecting emission lines in slitless spectroscopic data, enabling the study of extreme emission line galaxies and their evolution across cosmic time.

Contribution

The paper presents a novel automated line search technique using Bayesian statistics for slitless spectroscopy data, improving detection sensitivity and probabilistic redshift estimation.

Findings

Detected emission lines down to near noise levels in 3D-HST data.

Found that extreme emission line galaxies are nearly 10 times more common at z~1.5 than at lower redshifts.

Demonstrated the method's potential for future large-scale spectroscopic surveys.

Abstract

The multiplexing capability of slitless spectroscopy is a powerful asset in creating large spectroscopic datasets, but issues such as spectral confusion make the interpretation of the data challenging. Here we present a new method to search for emission lines in the slitless spectroscopic data from the 3D-HST survey utilizing the Wide-Field Camera 3 on board the Hubble Space Telescope. Using a novel statistical technique, we can detect compact (extended) emission lines at 90% completeness down to fluxes of 1.5 (3.0) times 10^{-17} erg/s/cm^2, close to the noise level of the grism exposures, for objects detected in the deep ancillary photometric data. Unlike previous methods, the Bayesian nature allows for probabilistic line identifications, namely redshift estimates, based on secondary emission line detections and/or photometric redshift priors. As a first application, we measure the comoving number density of Extreme Emission Line Galaxies (restframe [O III] 5007 equivalent widths in excess of 500 Angstroms). We find that these galaxies are nearly 10 times more common above z~1.5 than at z<0.5. With upcoming large grism surveys such as Euclid and WFIRST as well as grisms featuring prominently on the NIRISS and NIRCam instruments on James Webb Space Telescope, methods like the one presented here will be crucial for constructing emission line redshift catalogs in an automated and well-understood manner.