Cosmic Star-Formation Activity at z = 2.2 Probed by H α Emission-Line Galaxies

Abstract

Abstract We present a pilot narrow-band survey of H$\alpha$ emitters at z$=$ 2.2 in the Great Observatories Origins Deep Survey North (GOODS-N) field with MOIRCS instrument on the Subaru Telescope. The survey reached a 3$\ \sigma$ limiting magnitude of 23.6 (NB209), which corresponds to a 3$\ \sigma$ limiting line flux of 2.5 $\times$ 10$^{-17}\ $erg s$^{-1}\ $cm$^{-2}$ over a 56 arcmin$^2$ contiguous area (excluding a shallower area). From this survey, we have identified eleven H$\alpha$ emitters and one AGN at z$=$ 2.2 on the basis of narrow-band excesses and photometric redshifts. We obtained spectra for seven out of the new objects, including one AGN; also, an emission line above 3$\ \sigma$ was detected from all of them. We estimated star-formation rates (SFR) and stellar masses ($M_{\mathrm{star}}$) for individual galaxies. The average SFR and $M_{\mathrm{star}}$ are 27.8$\ M_{\odot}\ $yr$^{-1}\ $ and 4.0 $\times$ 10$^{10}\ M_\odot$, respectively. Their specific star-formation rates negatively correlate with their stellar masses. Fitting to a Schechter function yields the H$\alpha$ luminosity function with log$L$$=$ 42.82, log$\phi$$=$$-$2.78, and $\alpha$$=$$-$1.37. The average star-formation rate density in the survey volume is estimated to be 0.31$\ M_{\odot}\ $yr$^{-1}\ $Mpc$^{-3}$ according to the Kennicutt relation between the H$\alpha$ luminosity and the star-formation rate. We compared our H$\alpha$ emitters at z$=$ 2.2 in GOODS-N with narrow-band line emitters in other fields and clusters to see their time evolution and environmental dependence. We found that the star-formation activity is rapidly reduced from z$=$ 2.5 to z$=$ 0.8 in the cluster environment, while it only moderately changed in the field environment. This result suggests that the time scale of galaxy formation differs among different environments, and the star-forming activities in high density regions eventually overtake those in lower-density regions as a consequence of “galaxy-formation bias” at high redshifts.