Spectroscopic confirmation of quiescent galaxies at z\sim 2

Quiescent Galaxies in the 3D-HST Survey: Spectroscopic confirmation of a large number of galaxies with relatively old stellar populations at

Katherine E. Whitaker11affiliation: Astrophysics Science Division, Goddard Space Flight Center, Code 665, Greenbelt, MD 20771, USA 88affiliation: NASA Postdoctoral Program Fellow , Pieter G. van Dokkum22affiliation: Department of Astronomy, Yale University, New Haven, CT 06520, USA , Gabriel Brammer33affiliation: European Southern Observatory, Alonso de Córdova 3107, Casilla 19001, Vitacura, Santiago, Chile , Ivelina G. Momcheva22affiliation: Department of Astronomy, Yale University, New Haven, CT 06520, USA , Rosalind Skelton22affiliation: Department of Astronomy, Yale University, New Haven, CT 06520, USA , Marijn Franx44affiliation: Sterrewacht Leiden, Leiden University, NL-2300 RA Leiden, The Netherlands , Mariska Kriek55affiliation: Department of Astronomy, University of California, Berkeley, CA 94720, USA , Ivo Labbé44affiliation: Sterrewacht Leiden, Leiden University, NL-2300 RA Leiden, The Netherlands , Mattia Fumagalli44affiliation: Sterrewacht Leiden, Leiden University, NL-2300 RA Leiden, The Netherlands , Britt F. Lundgren66affiliation: Department of Astronomy, University of Wisconsin, Madison, WI 53706, USA , Erica J. Nelson22affiliation: Department of Astronomy, Yale University, New Haven, CT 06520, USA , Shannon G. Patel44affiliation: Sterrewacht Leiden, Leiden University, NL-2300 RA Leiden, The Netherlands , Hans-Walter Rix77affiliation: Max Planck Institut fur Astronomie, Königstuhl 17, D-69117 Heidelberg, Germany
Abstract

Quiescent galaxies at have been identified in large numbers based on rest-frame colors, but only a small number of these galaxies have been spectroscopically confirmed to show that their rest- frame optical spectra show either strong Balmer or metal absorption lines. Here, we median stack the rest-frame optical spectra for 171 photometrically-quiescent galaxies at from the 3D-HST grism survey. In addition to H (4861), we unambiguously identify metal absorption lines in the stacked spectrum, including the G-band (4304), Mg I (5175), and Na I (5894). This finding demonstrates that galaxies with relatively old stellar populations already existed when the universe was Gyr old, and that rest-frame color selection techniques can efficiently select them. We find an average age of Gyr when fitting a simple stellar population to the entire stack. We confirm our previous result from medium-band photometry that the stellar age varies with the colors of quiescent galaxies: the reddest 80% of galaxies are dominated by metal lines and have a relatively old mean age of Gyr, whereas the bluest (and brightest) galaxies have strong Balmer lines and a spectroscopic age of Gyr. Although the spectrum is dominated by an evolved stellar population, we also find [OIII] and H emission. Interestingly, this emission is more centrally concentrated than the continuum with erg s, indicating residual central star formation or nuclear activity.

Subject headings:
galaxies: evolution — galaxies: formation — galaxies: high-redshift
slugcomment: Accepted for publication in the Astrophysical Journal Letters

1. Introduction

In the nearby universe, the most massive galaxies almost always have little ongoing star-formation and red colors that exhibit a remarkably small intrinsic scatter (e.g., Bower et al., 1992). These red, quiescent galaxies form a well-defined color-mass relation known as the “red sequence”. Photometric studies of large, representative samples of galaxies based on broad-band and medium-band photometry have pushed the detection of the red sequence out to (e.g., Williams et al., 2009; Whitaker et al., 2011; Brammer et al., 2011; Nicol et al., 2011). However, only a small fraction of these distant quiescent galaxies have been spectroscopically confirmed (e.g., Cimatti et al., 2004; Daddi et al., 2005; Kriek et al., 2006; Cimatti et al., 2008; Kriek et al., 2008, 2009; van Dokkum & Brammer, 2010; van de Sande et al., 2011, 2012; Onodera et al., 2012; Toft et al., 2012; Bezanson et al., 2013).

Interestingly, most distant quiescent galaxies with high quality rest-frame optical spectra appear to exhibit young ages, showing strong Balmer absorption lines (e.g., van de Sande et al., 2012; Bezanson et al., 2013). This could imply that the bulk of high redshift quiescent galaxies were recently quenched. However, most spectroscopic studies in the literature are biased towards the brightest (and consequently youngest) of such galaxies. Therefore, spectroscopic ages for a representative sample of quiescent galaxies remains elusive.

Owing to the near-infrared (NIR) slitless spectroscopic capabilities provided by the Wide-Field Camera 3 (WFC3) on the Hubble Space Telescope (HST), it is now possible to obtain low-resolution spectroscopy of a mass-limited sample of quiescent galaxies at . In this letter, we stack the spectra of 171 quiescent galaxies at and demonstrate that they have absorption features indicative of evolved stellar populations. Furthermore, with ages derived from the stacked grism spectra, we are in a unique position to test what drives the color spread of quiescent galaxies for the first time.

We assume a CDM cosmology with =0.3, =0.7, and =70 km s Mpc throughout the paper. All magnitudes are given in the AB system.

2. Data

The 3D-HST treasury program (Brammer et al., 2012a), a 248–orbit NIR spectroscopic survey with the HST/WFC3 G141 grism, provides spatially resolved low-resolution spectra of all objects in five well-studied extra-galactic fields to a depth for continuum magnitudes of . 3D-HST has targeted the AEGIS, COSMOS, GOODS-S and UDS fields, as well as incorporated publicly-available data in the GOODS-N field (GO:11600; PI:Weiner). WFC3 imaging for all five fields is available from the CANDELS survey (Grogin et al., 2011; Koekemoer et al., 2011). With a wavelength range of 1.10m1.65m, the prominent age-dependent H4861 and Mg5175 absorption features in quiescent galaxies at are observable. The first-order dispersion of the G141 grism is 46 /pixel () with a spatial resolution of 0.12, sampled with 0.06 pixels; as the spectra have high spatial resolution and low spectral resolution, the line width almost exclusively reflects the size of the galaxy in the dispersion direction.

The sample used in this letter is selected from WFC3-selected photometric catalogs generated from the HST ACS and WFC3 images of the CANDELS and 3D-HST survey fields, along with ancillary ground-based optical and NIR images and mid-IR images with Spitzer/IRAC using a methodology similar to that described by Whitaker et al. (2011). A full description of these catalogs is beyond the scope of this letter and will be described in a forthcoming paper (Skelton et al. in prep). We extract a two-dimensional flux-calibrated WFC3/G141 grism spectrum for every object in the photometric catalog with . Details of the reduction and extraction of the grism spectra, including accounting for the contamination of overlapping objects, are provided by Brammer et al. (2012a, b).

To determine the galaxy redshifts, we first compute a purely photometric redshift from the photometry, using the EAZY code (Brammer et al., 2008). We then fit the full two-dimensional grism spectrum separately with a combination of the continuum template taken from the EAZY fit and a single emission-line-only template with fixed line ratios taken from the SDSS composite star-forming galaxy spectrum of Dobos et al. (2012). The final grism redshift, z_grism, is determined on a finely-sampled redshift grid with the photometry-only redshift probability distribution function used as a prior. This method is somewhat more flexible than that originally described by Brammer et al. (2012a), but the redshift precision is similar with . Finally, rest-frame colors and stellar population parameters are computed from the photometry with the EAZY and FAST (Kriek et al., 2009) codes, respectively, with the redshift fixed to z_grism and assuming a Chabrier (2003) initial mass function.

3. Sample Selection

A standard method for discriminating high redshift quiescent galaxies from star-forming galaxies is selecting on the rest-frame and colors (e.g., Labbé et al., 2005; Wuyts et al., 2007; Williams et al., 2009; Bundy et al., 2010; Cardamone et al., 2010; Whitaker et al., 2011; Brammer et al., 2011; Patel et al., 2012); quiescent galaxies have strong Balmer/4000Å breaks, characterized by red colors and relatively blue colors.

Following the definition of Whitaker et al. (2012), our quiescent selection box is shown in Figure 1, with the larger 3D-HST parent sample at shown in grey-scale. Using the dotted line in Figure 1 (), we further divide our quiescent sample into “younger” (blue) and “older” (red) galaxies (Section 5). The nature of the stellar populations of galaxies with rest-frame colors close to the quiescent/star-forming division is not clear and we therefore restrict our analysis to be more conservative by 0.08 mag (excluding those galaxies with high transparency in Figure 1). We select galaxies at with stellar masses log(M M, further restricting our sample to require wavelength coverage of the grism spectrum with of the pixels flagged as bad. The grism spectra of quiescent galaxies do not have strong emission line features and therefore the redshifts may not be reliable beyond certain magnitude limits; we additionally require that we detect the continuum at per resolution element in the 1D spectrum (), removing the faintest 20% of galaxies.

Figure 1.— Rest-frame and color selection of quiescent galaxies, further separating the sample into two bins of younger (blue) and older (red) quiescent galaxies in Section 5. All galaxies with are shown in grayscale.

4. Quiescent Stacks

Although the individual galaxies are too faint to discern spectral features, by stacking the spectra of our final sample of 171 galaxies we can achieve the necessary signal-to-noise ratio to robustly identify absorption features in a large well-defined high redshift sample for the first time. To normalize the spectra, we shift to the rest-frame and interpolate the flux values to a new wavelength grid with 10 width bins. Next, we fit the continuum in each spectrum with a third order polynomial, masking regions around prominent absorption features. After dividing by the continuum, we determine the median flux value at each wavelength bin and smooth with a boxcar of 20. The grey-scale error bars are derived from 100 bootstrap iterations of the stacking analysis.

We show the stacked spectrum in Figure 2. In the top panel, we see many clear absorption features: G-band () blended with H(), Mg I (), Fe I (), and Na I (). These metal lines are indicative of an evolved stellar population. We caution that systematic residuals in the blue part of the spectrum are of similar strength as the emission lines. The residuals do not seem to be caused by photon noise, and show up in all stacked spectra. We have not identified the cause of these residuals. We cannot exclude errors in the stellar population synthesis models, although that is unlikely in this well-studied wavelength range.

Figure 2.— Median rest-frame stack of 171 quiescent galaxies (top) and the residuals after subtracting the best-fit model (red) with an age of 1.25 Gyr (bottom), using the grism redshifts. The orange model shows an additional 5% enhancement of centrally-concentrated H and [OIII] emission. The accuracy of the grism redshifts enables us to resolve absorption features at with high signal-to-noise for the first time.

We perform a least-squares minimization using the Vazdekis et al. (2010) models to find the best-fit age of the stack. The Vazdekis et al. (2010) models provide moderately high resolution SEDs computed assuming an instananeous burst, Solar metallicity and a Kroupa (2001) universal initial mass function, based on the MILES (Sánchez-Blázquez et al., 2006) stellar library. We simulate the expected morphological broadening by convolving the high-resolution models with the grism response and the object morphology in the direct image (see Brammer et al., 2012b, for more information on how the 2-D model spectra are generated). One-dimensional model spectra are then extracted in the same way as was done for the observed G141 spectra. The red line in Figure 2 is the best-fit convolved model with an age of 1.25 Gyr, fit to all wavelengths.

Figure 3.— The morphology of the residual [OIII] emission after subtracting a 1.25 Gyr model (histogram). The morphology is compared to the expectation for a centrally-concentrated gas distribution (red line) and to a distribution that follows the galaxy light (black line). The red shaded region indicates the confidence interval resulting from the modeling analysis. The resolved double-peak of the [OIII] feature is only expected for centrally-concentrated emission with the low grism resolution.

The bottom panel of Figure 2 shows the residuals when subtracting the best-fit model (red). Here, we see the [OIII] doublet () and H in emission. It is somewhat surprising to clearly resolve the [OIII] doublet given the morphological broadening in the grism spectra. In Figure 3, we demonstrate the expected resolution for the [OIII] emission feature at a fixed line ratio of 3:1 when convolving with the interlaced PSF (red) and the average galaxy profile (black). The stacked profile is constructed by re-centering, masking neighbors and normalizing by the total flux of individual galaxies before taking the mean/median. The [OIII] doublet is blended together when the emission is a global feature of the galaxy, whereas centrally-concentrated emission (within roughly the central pixel) shows two clear peaks. The point source nature of the [OIII] emission lines also indicates that the high quality of the grism redshifts, as redshift errors would broaden the observed [OIII] lines in the stack. We note that although the relative ratio of the [OIII] doublet is mildly sensitive to the assumed best-fit model, this effect is significantly less than the error bars from the stacking analysis themselves. The residuals are consistent with both centrally-concentrated [OIII] and H emission.

To model the central H and [OIII] enhancements, we simultaneously fit a grid of Vazdekis et al. (2010) models convolved with the average galaxy profile with enhancements ranging from 0–15% above the continuum flux for H and [OIII] emission lines convolved with the PSF. We assume a fixed line ratio of , consistent with the residuals in Figure 2 and line ratio diagnostics for post-starburst galaxies in the Sloan Digital Sky Survey (SDSS; Mendel et al., 2013). The best-fit model of 1.25 Gyr with a 5% [OIII] and H central enhancement is shown in the top panel of Figure 2 (orange).

We measure the [OIII] line flux from the residuals in Figure 2 to be erg s cm, which corresponds to of erg s for the median redshift of 1.64. Similarly, we measure the H line flux to be erg s cm. To determine the line flux errors, we add the simulation errors in quadrature to the error in the average continuum level. The resulting [OIII]/H ratio is . If we further correct for the average dust extinction value of determined from the photometry, we estimate the dust-corrected [OIII] luminosity to be erg s, following Bassani et al. (1999). Adopting the conversion from to from Lamastra et al. (2009) and a factor of ten conversion from to (Lusso et al., 2010), we estimate a typical active galactic nucleus (AGN) luminosity to be erg s. However, we note that it is not possible to differentiate between centrally-concentrated residual star-formation or an AGN without further information.

5. What drives the spread in rest-frame colors?

Figure 4.— Median stacks of quiescent galaxies in two rest-frame color bins (as defined in Figure 1). The top two panels show the number of galaxies and median redshift of each wavelength bin. The observed stacks in the third panel reveal that the bluest quiescent galaxies have younger ages, most notable by the stronger H and weaker Mg I absorption. The fourth panel shows the best-fit models, including a 5–6% centrally-concentrated enhancement of H and [OIII] emission. The residuals are shown in the bottom panel.

With these high quality stacked spectra, we can now begin to assess the driving factor behind the spread in rest-frame colors of quiescent galaxies. The three possibilities include age, dust or metallicity variations amongst galaxy populations. The difference in rest-frame and colors corresponding to 1 Gyr of passive aging within the Bruzual & Charlot (2003) models results in a 0.4 mag vector parallel to the quiescent sequence. Similarly, increasing the metallicity from log(Z)=0.02 (Solar) to log(Z)=0.05 or adding 0.5 mag of dust reddening following the Calzetti et al. (2000) dust law results in an almost identical vector in both direction and magnitude. Due to degeneracies between age, dust and metallicity, it is notoriously difficult to differentiate which of these effects dominates.

Metallicity variations will cause the color spread in galaxies to increase over time as the galaxies passively evolve, reaching a roughly constant scatter for the oldest stellar populations. As noted by Whitaker et al. (2010), the trend for the intrinsic scatter in the rest-frame colors of massive quiescent galaxies to steadily decrease from to the present epoch (see also, e.g., Ruhland et al., 2009; Papovich et al., 2010) is opposite to that predicted from metallicity variations alone. We therefore expect that metallicity will not have a large effect at high redshifts.

As the shape of the Balmer/4000Å break depends sensitively on the treatment of dust reddening, one would ideally need to correlate direct, independent measurements of the ages of massive galaxies with their colors. Surprisingly, Bezanson et al. (2013) did not find a simple correlation between the rest-frame color and the strength of the Balmer absorption lines, as some of the reddest galaxies in the spectroscopic sample had strong Balmer absorption lines. However, with a small sample size of 13 galaxies, it is difficult to gauge if these results are representative of the overall quiescent population.

In Figure 4, we plot the median stacked spectra in two rest-frame color bins for the 171 quiescent galaxies, as color-coded in Figure 1. The total number of galaxies in each bin is labeled in Figure 1 and the number of galaxies per wavelength bin is indicated in the top histogram of Figure 4. A minimum number of 20 galaxies was required per wavelength bin to ensure a robust measurement.

Here, we assess the effects of age as a function of rest-frame color for a mass-limited sample of quiescent galaxies by measuring the strength of absorption features such as Mg I and H. Both of these spectral features are sensitive to the age of the stellar population, but not very sensitive to metallicity or dust reddening. Given the narrow baseline of the spectral features, dust extinction is not expected to have a significant effect. At the grism resolution, the H absorption feature is expected to be about 4–5% stronger in a 1 Gyr stellar population as compared to a 2 Gyr stellar population, whereas the Mg I absorption feature is about 3–4% weaker. Increasing the metallicity results in a slightly deeper Mg I feature, but only on the order of . To first order, any change measured in H and Mg I is due to age differences between the galaxies.

The average age of galaxies with the bluest rest-frame colors is Gyr, consistent with the expectation that they are in the “post-starburst” phase (Kriek et al., 2010; Whitaker et al., 2010, 2012). These galaxies clearly show significantly stronger H and H absorption than the reddest galaxies. Note the difference in shape of the blended G-band and H between these two stacks. The redder galaxies are 0.7 Gyr older on average, with a best-fit age of Gyr. For a 0.7 Gyr age difference, the Bruzual & Charlot (2003) models predict and , corresponding to exactly the observed difference in the median rest-frame colors of the two bins.

The older quiescent galaxies show clear signs of either central residual star-formation activity or an AGN, while the enhancement in the younger galaxies is not well-constrained despite a well-determined spectroscopic age. We measure the residual [OIII] and H line fluxes for the younger and older populations, finding ratios of and respectively. The bluer galaxies lack a clear [OIII] feature, although the difference with the redder galaxies is only marginally significant.

6. Discussion

We present stacked grism spectroscopy of a mass-limited sample of 171 quiescent galaxies at from the 3D-HST treasury program (Brammer et al., 2012a). We demonstrate that we can resolve absorption features such as the G-band (4304), H4861), Mg I (5175), and Na I (5894) in the median stacks due to the high quality grism redshifts. We further detect centrally-concentrated H and [OIII] in emission, indicating either residual central star-formation (e.g., Bezanson et al., 2013) or an AGN.

The average L derived from the stacks of L is consistent with the results of Mendel et al. (2013), who find that 70-80% of recently-quenched galaxies in the SDSS have LINER-like active nuclei (L L). Furthermore, the [OIII]/H ratio of 0.9 for a median mass of M places the average quiescent galaxy within the LINER region of the Mass-Excitation diagnostic diagram (Juneau et al., 2011), albeit marginally consistent with the “composite” crossover region within the error bars. From the average dust-corrected [OIII] (5007) luminosity, we infer that quiescent galaxies may have a typical AGN luminosity of erg s. These results are consistent with the recent work of Olsen et al. (2013), who find that 70–100% of massive quiescent galaxies at contain a low- or high-luminosity AGN. The potential ubiquitous nature of AGNs in massive, quiescent galaxies provides observational support that black hole accretion may be more effective at the high-mass end (e.g., Kriek et al., 2007).

To understand the relationship between rest-frame color and our spectroscopic age measurements, we further stack the data in two rest-frame color bins. We find that the bluest quiescent galaxies have a more prominent H absorption line, with best-fit ages of Gyr, confirming the idea that these galaxies are the most recently quenched. Furthermore, we find that redder quiescent galaxies (80% of the population) have an older age of Gyr. Although previous spectroscopic studies have measured similarly old ages (e.g., Kriek et al., 2009; Onodera et al., 2012; van de Sande et al., 2012), this is the first time we are able to probe the full population. These results suggest that age varies significantly with the colors of quiescent galaxies, consistent with Whitaker et al. (2010).

Galaxies with the reddest rest-frame colors likely have some degree of dust extinction, as dust-free models cannot produce such red colors at these redshifts. Dust extinction does not greatly affect the line-index measurements for single stellar populations given the narrow baseline of the spectral features, but its effect can be significant for the 4000Å break (MacArthur, 2005). Consequently, although the absorption features of the 3D-HST stacks will be effectively independent of dust extinction, the rest-frame color can change significantly in the presence of dust due to its sensitivity to the continuum shape. We cannot rule out that the reddest galaxies have a potentially significant contribution from younger, dusty quiescent galaxies. Nonetheless, despite this caveat, we measure a significant age difference between the two populations of 0.7 Gyr that is consistent with the observed rest-frame color differences.

The key result from this paper is that we have conclusively demonstrated that old, quiescent galaxies exist in large numbers at . This is not a surprise given extensive previous results that were based on broadband and medium-band photometry (e.g., Brammer et al., 2011), and ground-based spectroscopy for small samples (e.g., Onodera et al., 2012), but our detection of metal lines in the stacked spectrum puts any lingering concerns to rest. Our result also implies that the “UVJ technique”, first described in Labbé et al. (2005), is very effective in identifying quiescent galaxies at .

We sincerely thank the referee for the thorough review of the manuscript. This research was supported by an appointment to the NASA Postdoctoral Program at the Goddard Space Flight Center, administered by Oak Ridge Associated Universities through a contract with NASA. Support from STScI grant GO-12177 and NASA ADAP grant NNX11AB08G is gratefully acknowledged.

References

  • Bassani et al. (1999) Bassani, L., Dadina, M., Maiolino, R., et al. 1999, ApJS, 121, 473
  • Bezanson et al. (2013) Bezanson, R., van Dokkum, P., van de Sande, J., et al. 2013, ApJ, 764, L8
  • Bower et al. (1992) Bower, R. G., Lucey, J. R., & Ellis, R. S. 1992, MNRAS, 254, 601
  • Brammer et al. (2012a) Brammer, G. B., van Dokkum, P. G., Franx, M., et al. 2012b, ApJS, 200, 13
  • Brammer et al. (2012b) Brammer, G. B., Sánchez-Janssen, R., Labbé, I., et al. 2012a, ApJ, 758, L17
  • Brammer et al. (2008) Brammer, G. B., van Dokkum, P. G., & Coppi, P. 2008, ApJ, 686, 1503
  • Brammer et al. (2011) Brammer, G. B., Whitaker, K. E., van Dokkum, P. G., et al. 2011, ApJ, 739, 24
  • Bruzual & Charlot (2003) Bruzual, G., & Charlot, S. 2003, MNRAS, 344, 1000
  • Bundy et al. (2010) Bundy, K., Scarlata, C., Carollo, C. M., et al. 2010, ApJ, 719, 1969
  • Calzetti et al. (2000) Calzetti, D., Armus, L., Bohlin, R. C., et al. 2000, ApJ, 533, 682
  • Cardamone et al. (2010) Cardamone, C. N., Urry, C. M., Schawinski, K., et al. 2010, ApJ, 721, L38
  • Chabrier (2003) Chabrier, G. 2003, PASP, 115, 763
  • Cimatti et al. (2008) Cimatti, A., Cassata, P., Pozzetti, L., et al. 2008, A&A, 482, 21
  • Cimatti et al. (2004) Cimatti, A., Daddi, E., Renzini, A., et al. 2004, Nature, 430, 184
  • Daddi et al. (2005) Daddi, E., Renzini, A., Pirzkal, N., et al. 2005, ApJ, 626, 680
  • Dobos et al. (2012) Dobos, L., Csabai, I., Yip, C.-W., et al. 2012, MNRAS, 420, 1217
  • Grogin et al. (2011) Grogin, N. A., Kocevski, D. D., Faber, S. M., et al. 2011, ApJS, 197, 35
  • Juneau et al. (2011) Juneau, S., Dickinson, M., Alexander, D. M., et al. 2011, ApJ, 736, 104
  • Koekemoer et al. (2011) Koekemoer, A. M., Faber, S. M., Ferguson, H. C., et al. 2011, ApJS, 197, 36
  • Kriek et al. (2010) Kriek, M., Labbé, I., Conroy, C., et al. 2010, ApJ, 722, L64
  • Kriek et al. (2008) Kriek, M., van der Wel, A., van Dokkum, P. G., et al. 2008, ApJ, 682, 896
  • Kriek et al. (2007) Kriek, M., van Dokkum, P. G., Franx, M., et al. 2007, ApJ, 669, 776
  • Kriek et al. (2006) —. 2006, ApJ, 649, L71
  • Kriek et al. (2009) Kriek, M., van Dokkum, P. G., Labbé, I., et al. 2009, ApJ, 700, 221
  • Kroupa (2001) Kroupa, P. 2001, MNRAS, 322, 231
  • Labbé et al. (2005) Labbé, I., Huang, J., Franx, M., et al. 2005, ApJ, 624, L81
  • Lamastra et al. (2009) Lamastra, A., Bianchi, S., Matt, G., et al. 2009, A&A, 504, 73
  • Lusso et al. (2010) Lusso, E., Comastri, A., Vignali, C., et al. 2010, A&A, 512, A34
  • MacArthur (2005) MacArthur, L. A. 2005, ApJ, 623, 795
  • Mendel et al. (2013) Mendel, J. T., Simard, L., Ellison, S. L., et al. 2013, MNRAS, 429, 2212
  • Nicol et al. (2011) Nicol, M.-H., Meisenheimer, K., Wolf, C., et al. 2011, ApJ, 727, 51
  • Olsen et al. (2013) Olsen, K. P., Rasmussen, J., Toft, S., et al. 2013, ApJ, 764, 4
  • Onodera et al. (2012) Onodera, M., Renzini, A., Carollo, M., et al. 2012, ApJ, 755, 26
  • Papovich et al. (2010) Papovich, C., Momcheva, I., Willmer, C. N. A., et al. 2010, ArXiv e-prints
  • Patel et al. (2012) Patel, S. G., Holden, B. P., Kelson, D. D., et al. 2012, ApJ, 748, L27
  • Ruhland et al. (2009) Ruhland, C., Bell, E. F., Häußler, B., et al. 2009, ApJ, 695, 1058
  • Sánchez-Blázquez et al. (2006) Sánchez-Blázquez, P., Peletier, R. F., Jiménez-Vicente, J., et al. 2006, MNRAS, 371, 703
  • Toft et al. (2012) Toft, S., Gallazzi, A., Zirm, A., et al. 2012, ApJ, 754, 3
  • van de Sande et al. (2012) van de Sande, J., Kriek, M., Franx, M., et al. 2012, ArXiv e-prints
  • van de Sande et al. (2011) —. 2011, ApJ, 736, L9+
  • van Dokkum & Brammer (2010) van Dokkum, P. G., & Brammer, G. 2010, ApJ, 718, L73
  • Vazdekis et al. (2010) Vazdekis, A., Sánchez-Blázquez, P., Falcón-Barroso, J., et al. 2010, MNRAS, 404, 1639
  • Whitaker et al. (2012) Whitaker, K. E., Kriek, M., van Dokkum, P. G., et al. 2012, ApJ, 745, 179
  • Whitaker et al. (2011) Whitaker, K. E., Labbé, I., van Dokkum, P. G., et al. 2011, ApJ, 735, 86
  • Whitaker et al. (2010) Whitaker, K. E., van Dokkum, P. G., Brammer, G., et al. 2010, ApJ, 719, 1715
  • Williams et al. (2009) Williams, R. J., Quadri, R. F., Franx, M., et al. 2009, ApJ, 691, 1879
  • Wuyts et al. (2007) Wuyts, S., Labbé, I., Franx, M., et al. 2007, ApJ, 655, 51
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
255308
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description