|
Using the Gemini Near-Infrared Spectrograph (GNIRS), astronomers have obtained an ultra-deep (~29 hour integration) near-infrared spectrum of a massive compact quiescent galaxy at a redshift of about 2.2 (when the universe was only about three billion years old). Understanding how a massive galaxy–especially this early in the history of the universe–can squelch star birth with such efficiency is a challenging problem for both observational and theoretical astrophysicists. The team, led by Mariska Kriek of Princeton University, used GNIRS on Gemini South to dissect the light of the galaxy, identified as 1255-0, in what is the deepest single-slit near infrared spectrum ever taken of a galaxy with a redshift greater than 2 (Figure 1). The spectrum, for the first time, shows absorption line features in this type of galaxy that allow for accurate redshift determination (Figure 2) and is providing insights into this unique category of galaxies that are very massive yet display extremely efficient squelching of star formation at this epoch. See Kriek et al.’s earlier work on this subject (also using GNIRS) here. A general dearth of spectroscopic data on galaxies of this type and redshift (age) make it very difficult to piece together an evolutionary scenario that explains how these galaxies can so quickly go from what must have been a remarkably frenzied period of star birth in the early universe to the quiescence phase that we see at redshifts of around 2. The Gemini GNIRS data, thanks to the detection of stellar continuum emission and of faint emission lines (ratios obtained for [NII]/H(alpha), [SII/H(alpha) and [OII]/[OIII]), allow analysis of star formation which is estimated to be less than 1% of the galaxy’s past average star formation rate. The emission-line ratios are also typical of Low-Ionization Nuclear Emission-line Region (LINER) galaxies. Best fit models to the spectrum imply an age of formation corresponding to a redshift range of z = 4 to 7. This work has been published in The Astrophysical Journal, 700:221-231, July 20, 2009. The paper is titled: "AN ULTRA-DEEP NEAR-INFRARED SPECTRUM OF A COMPACT QUIESCENT GALAXY AT z = 2.2," and the authors are: Mariska Kriek, Pieter G. van Dokkum, Ivo Labbé, Marijn Franx, Garth D. Illingworth, Danilo Marchesini, & Ryan F. Quadri. |
Figure 1: The ~ 29 hrs deep GNIRS spectrum of 1255-0 (black). The spectrum is sampled to bins of 50 °A in observed frame, and smoothed by a boxcar of 75 Å. The gray shaded background indicates the noise level of the spectrum, with dark being noisier. We omitted the regions in between the atmospheric windows. Overplotted in green is the best-fit Bruzual & Charlot (2003) SPS model to the spectrum and optical-IR photometry, assuming a Chabrier (2003) IMF and solar metallicity. This fit corresponds to a stellar mass of 2.3 x 1011M⊙, an age of 2.1 Gyr, a τ of 0.3 Gyr, a reddening of Av = 0.25 mag, a SFR of 1.9 M⊙/yr, and a specific SFR of 0.008 Gyr−1. The locations of possible absorption and emission lines are indicated by the red dashed and blue dotted lines, respectively. 
Figure 2: Reduced χ2 vs. continuum redshift (number of degrees of freedom Ndof = 319), when leaving redshift as a free parameter in the SPS model (Bruzual & Charlot 2003) fitting. The other parameters, including metallicity are allowed to vary. The rest-frame UV-NIR broadband photometry is not included in the fit. The 68% and 95% confidence intervals are indicated by the dotted lines. The best-fit emission line redshift is indicated by the vertical solid line, and its 68% and 95% confidence intervals by the green and blue shaded regions, respectively. The continuum emission yields a well constrained spectroscopic redshift. The redshift difference between the stellar and nebular emission is not at all significant. |