Metallicity and SFR coevolution

Coevolution of metallicity and star formation in galaxies to . II. A theoretical model

Leslie Hunt, Pratika Dayal, Laura Magrini and Andrea Ferrara
INAF/Osservatorio Astrofisico di Arcetri, Largo Enrico Fermi 5, 50125 Firenze, Italy
Kapteyn Astronomical Institute, University of Groningen, P.O. Box 800, 9700 AV Groningen, The Netherlands
Scuola Normale Superiore, Piazza dei Cavalieri 7, I-56126 Pisa, Italy
draft version 8 August 2016

Recent work suggests that galaxy evolution, and the build-up of stellar mass (M) over cosmic time, is characterized by changes with redshift of star formation rate (SFR) and oxygen abundance (O/H). In a companion paper, we have compiled a large dataset to study Metallicity Evolution and Galaxy Assembly (MEGA), consisting of 1000 galaxies to with a common O/H calibration. Here we interpret the MEGA scaling relations of M, SFR, and O/H with an updated version of the model presented by Dayal et al. (2013). This model successfully reproduces the observed O/H ratio of 80 000 galaxies selected from the Sloan Digital Sky Survey to within 0.050.06 dex. By extending the model to the higher redshift MEGA sample, we find that although the specific mass loading of outflows does not change measurably during the evolution, the accretion rate and gas content of galaxies increase significantly with redshift. These two effects can explain, either separately or possibly in tandem, the observed lower metal abundance of high- galaxies.

galaxies: evolution – galaxies: abundances – galaxies: star formation – galaxies: high redshift
pagerange: Coevolution of metallicity and star formation in galaxies to . II. A theoretical modelReferencespubyear: 2016

1 Introduction

Galaxy evolution takes place through the build-up over time of stellar mass (M) through various episodes of star formation. As stars evolve, metals are produced, so that metallicity and its relation with M and star-formation rate (SFR) are important gauges of star-formation history. Thus, metal content, typically measured through the gas-phase oxygen abundance (O/H), the most abundant heavy element produced by massive stars, can be used to assess a galaxy’s evolutionary state. The integrated star-formation activity over the lifetime of the galaxy is measured by stellar mass, while the star-formation rate (SFR) measures how much gas is being converted into stars over the time scale of the SFR indicator. The gas-phase metallicity () quantifies the production of metals from high-mass stars, but also indicates the degree to which galaxies interact with their environment through gas accretion and outflows in the form of galactic winds.

Several well-established scaling relations reflect the intimate link between M, SFR, and O/H. A mass-metallicity relation (MZR) connecting stellar mass and is clearly present in the Local Universe (e.g., Tremonti et al., 2004), and it apparently extends to the highest redshifts examined so far (e.g., Erb et al., 2006a; Maiolino et al., 2008; Mannucci et al., 2009; Cresci et al., 2012; Xia et al., 2012; Yabe et al., 2012; Henry et al., 2013; Cullen et al., 2014; Zahid et al., 2014; Troncoso et al., 2014; Steidel et al., 2014; Wuyts et al., 2014; Ly et al., 2015; de los Reyes et al., 2015; Onodera et al., 2016). Stellar mass and SFR are also related in a SF “main sequence” (SFMS) both locally (Brinchmann et al., 2004; Salim et al., 2007) and at high redshift (e.g., Noeske et al., 2007; Elbaz et al., 2011; Karim et al., 2011; Wuyts et al., 2011; Speagle et al., 2014). It is generally agreed that the high-redshift relations of both the MZR and the SFMS are similar in form to the local ones, but they show different normalizations: at a given M, SFR (and sSFR) increases with increasing redshift while metallicity decreases. Moreover, both relations show an inflection at high M, with a decrease in O/H and SFR (Tremonti et al., 2004; Wyder et al., 2007; Noeske et al., 2007; Whitaker et al., 2014; Lee et al., 2015; Gavazzi et al., 2015).

In order to observationally constrain the evolution of metallicity with redshift, we have compiled a new dataset of star-forming galaxies from to with nebular oxygen abundance measurements; we have dubbed this compilation the “MEGA” dataset, corresponding to Metallicity Evolution and Galaxy Assembly. This dataset presents a marked improvement over the dataset used by Hunt et al. (2012) because of the inclusion of several more high samples and, more importantly, because of a common metallicity calibration. The coevolution of SFR and O/H with redshift in the MEGA dataset is presented in a companion paper (Hunt et al. 2016, hereafter Paper I), and summarized in Section 2. Here, we describe a physical model for understanding the coevolution of M, SFR, and O/H. This is an extension of the model presented by Dayal et al. (2013), and also a re-assessment of the model parameters with different O/H calibrations, different stellar yields, and a different Initial Mass Function (IMF). Section 3 describes this analytical approach to physically interpret the scaling relations locally, and Section 4 extends the analytical formulation to quantify evolution of the scaling relations with redshift. We discuss the results of our modeling in Sect. 5, and summarize our conclusions in Sect. 6. As in Paper I, we use a Chabrier (2003) IMF throughout.

2 Metallicity and SFR coevolution

Paper I presents common O/H calibrations for 990 galaxies from to culled from 19 different samples. The MEGA dataset spans a range of in M, in SFR, and almost two orders of magnitude in O/H. The conversion to common metallicity calibrations was performed according to the transformations given by Kewley & Ellison (2008). We chose three O/H calibrations, because of their similarity to electron-temperature or “direct” methods (e.g., Andrews & Martini, 2013): Denicoló et al. (2002, hereafter D02) and the nitrogen and oxygennitrogen-based abundances by Pettini & Pagel (2004, hereafter PP04N2 and PP04O3N2, respectively).

We hypothesized that the inflections in the MZR and SFMS at high M were compensating one another, and also that the decrease in O/H at high redshift was somehow quantitatively associated with the increase in SFR. This hypothesis was essentially aimed at assessing the possibility that the mutual correlations among M, SFR, and O/H could be expressed as a planar relation, i.e., with only two of the three pseudo-observables being required to describe the variation. From a Principal Component Analysis (PCA), we showed that this was indeed possible, and found a 2-dimensional relation, namely the Fundamental Plane of Metallicity (FPZ):


The FPZ is an accurate reflection of the relations among M, SFR, and O/H at least to within 0.16 dex in O/H (Eqn. (1) represents the best-behaved O/H calibration, PP04N2, Pettini & Pagel, 2004). As described in Paper I, such a level of residuals is comparable to trends found in other galaxy samples but with a smaller range in M, SFR, and O/H.

We also analyzed whether or not the FPZ varied over time, and found a significant correlation of the residuals with redshift, but with a residual standard error of 0.16 dex, the same as the 0.16 dex uncertainty of the FPZ itself. Thus, we concluded that the FPZ is approximately redshift invariant, since any redshift variation of the FPZ is within the noise of the current data.

As a comparison to the MEGA dataset, here and in Paper I we also analyze the emission-line galaxies from the Sloan Digital Sky Survey (SDSS) defined by Mannucci et al. (2010, hereafter the SDSS10 sample). Although its parameter-space coverage is relatively limited, because of its large size this sample adds superb statistical power to our analysis, and also enables a comparison with Dayal et al. (2013) who analyzed the same galaxies. As for the MEGA dataset, we have converted the (Kewley & Dopita, 2002, KD02) metallicities reported by Mannucci et al. (2010) to the D02, PP04N2, and PP04O3N2 calibrations, as described in Paper I.

3 Coevolution: physical insights

Much theoretical effort has been devoted to understand the evolution of the MZR and its dependence, if any, on SFR and the surrounding environment (e.g., Peng et al., 2010; Peeples & Shankar, 2011; Davé et al., 2011; Yates et al., 2012; Hopkins et al., 2012; Krumholz & Dekel, 2012; Dayal et al., 2013; Lilly et al., 2013; Schaye et al., 2015). Here, following the analytical formalism of Dayal et al. (2013), we present a physical basis for the observed coevolution of SFR and O/H. With a nebular oxygen abundance where and are the galaxy oxygen and total gas mass, respectively, SFR, O/H, and M are related through a set of evolutionary differential equations (see Dayal et al., 2013):


Here  is the star-formation timescale and we have renamed SFR as ; we assume that the SFR is proportional to the gas mass, and that the infalling gas is much less metal enriched than the ambient ISM, . The two constants, and , represent the returned fraction from stars and the yield per stellar generation, respectively, and depend on the IMF; we have adopted , taken from Vincenzo et al. (2016) using the stellar yields from Nomoto et al. (2013) assuming a Chabrier (2003) IMF111The stellar yields from Romano et al. (2010) are similar, with . Our values for are roughly the mean values of the yields over the range of metallicities given by Vincenzo et al. (2016) since our model does not contemplate time variation of the yields.. The lower (upper) mass limit for these yields is 0.1  (100 ).

Like Dayal et al. (2013), we further assume that gas outflow through winds, is proportional to the SFR, where is the total (darkbaryonic) galaxy mass. There is some observational evidence of such a proportionality, at least in the hot and/or ionized gas components of galaxies (e.g., Martin, 2005; Veilleux et al., 2005). For convenience in the analytical formulation, we also take gas accretion, , to be proportional to SFR. Although there is little evidence to directly support such a proportionality, it may be that gas is replenished in galaxies by cooling of pre-existing ionized gas in expanding supershells or galactic fountains (e.g., Hopkins et al., 2008; Fraternali & Binney, 2008; Fraternali & Tomassetti, 2012).

The solution of these equations can be written as:


where , and . The initial gas mass can be expressed as:


Altogether, there are five free parameters in the () model: the inverse of the star-formation timescale, ; and the coefficients and power-law indices describing the mass dependence of infall through gas accretion (), ; and of outflow by galactic winds (), .

The solution to these equations results in an implicit star-formation history (SFH) assumed by our models, namely an exponentially declining one. Such a SFH is observationally supported by some studies (e.g., Eyles et al., 2007; Stark et al., 2009), but contested by others (e.g., Pacifici et al., 2015; Salmon et al., 2015). In our case, the exponential time scale depends on halo mass (here assumed to be proportional to M). As described by Dayal et al. (2013), this means that galaxies of different masses build up their metal contents at different rates: galaxies start from low , low M, and high values and then move towards higher metallicities as their M increases. However, depending on their mass, they move at different velocities along the track: the most massive observed galaxies are very evolved with a large specific age (i.e., in units of the star formation timescale) and move essentially along constant metallicity curves; smaller objects are younger and are still gently building up their metal content.

We can accommodate this model (hereafter the FPZ model) to the MEGA dataset and the SDSS10 sample for the three O/H calibrations considered here by fitting for these five parameters. Fig. 1 illustrates the resulting best-fit FPZ model overplotted on the binned SDSS10 data for all three O/H calibrations (D02, PP04N2, PP04O3N2). Since this is a local calibration, the FPZ model fit to the MEGA dataset included only galaxies with ; Fig. 2 shows this best fit to the MEGA data set of the FPZ model. Table LABEL:tab:models reports the best-fit parameters.

The model is an excellent representation of the SDSS10 data (see also Dayal et al., 2013); over 80 000 individual data points, the best-fit model with only five free parameters gives a residual error in 12log(O/H) predicted vs. observed of 0.05–0.07 dex, according to the O/H calibration. The model is somewhat less successful at fitting the MEGA dataset, with mean O/H residuals of predicted vs. observed of  dex, larger than the FPZ residuals of  dex (see Paper I). However, these galaxies are only a subset of the entire MEGA sample (because of the redshift limitation), and the best-fit MEGA parameters are generally consistent, within the errors, to SDSS (see below for more details).

Figure 1: SDSS10 galaxies: 12log(O/H) plotted against (log of) SFR; the data points correspond to the average 12log(O/H) (with the three O/H calibrations) for SDSS galaxies in a given SFR bin, and the error bars to the spread of the data. The curves show three FPZ models with different O/H calibrations (D02, PP04N2, PP04O3N2) as discussed in the text. The color coding is by (log) M: 7.5-8 (magenta), 8-8.5 (purple), 8.5-9 (blue), 9-9.5 (cyan), 9.5-10 (green), 10-10.5 (orange), 10.5-11 (red), 11-11.5 (brown). The model and the data are in excellent agreement with a residual scatter of 0.050.07 dex, according to the O/H calibration (see Table LABEL:tab:models).

Figure 2: MEGA galaxies: 12log(O/H) plotted against (log of) SFR; the data points correspond to the average 12log(O/H) (with the three O/H calibrations) for galaxies in a given SFR bin, and the error bars to the spread of the data. The curves show three FPZ models with different O/H calibrations (D02, PP04N2, PP04O3N2) as discussed in the text; here we show the FPZ best-fit MEGA model, rather than the FPZ best-fit SDSS model given in Fig. 1. As in Fig. 1, the color coding is by (log) M: 7.5-8 (magenta), 8-8.5 (purple), 8.5-9 (blue), 9-9.5 (cyan), 9.5-10 (green), 10-10.5 (orange), 10.5-11 (red), 11-11.5 (brown). Only bins with more than 2 galaxies are shown; in some cases the standard deviations within the bins are smaller than the data point. The residual scatter of the model is 0.240.26 dex, according to the O/H calibration (see Table LABEL:tab:models).
Sample Degrees of O/H
freedom calibration (Gyr)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
Local Universe
MEGA 252 0.264 D02   
MEGA 252 0.243 PP04N2
MEGA 252 0.251 PP04O3N2
SDSS10 78536 0.052 D02  
SDSS10 78536 0.063 PP04N2
SDSS10 78536 0.068 PP04O3N2  
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
MEGA 25 0.060 D02  
MEGA 25 0.038 PP04N2
MEGA 25 0.046 PP04O3N2  
Schechter 28 0.079 D02  
Schechter 28 0.025 PP04N2
Schechter 28 0.025 PP04O3N2  
: ,
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
MEGA 25 0.046 D02    
MEGA 25 0.039 PP04N2
MEGA 25 0.049 PP04O3N2  
Schechter 28 0.035 D02    
Schechter 28 0.042 PP04N2
Schechter 28 0.049 PP04O3N2  
:  ]
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
MEGA 25 0.062 D02  
MEGA 25 0.039 PP04N2
MEGA 25 0.048 PP04O3N2  
Schechter 28 0.035 D02  
Schechter 28 0.026 PP04N2
Schechter 28 0.032 PP04O3N2  

 The free parameters in the FPZ models are: ; ; . At , there are three classes of best-fit models as described in the text: one that depends on the variation of the inverse of the SF timescale (); one that depends on the variations of the inflow () and outflow () amplitudes; and one that depends on , the relative gas content.

 Residual standard error of global fit.

 D02 corresponds to Denicoló et al. (2002); PP04N2 and PP04O3N2 to nitrogen-based and oxygen+nitrogen based calibrations by Pettini & Pagel (2004).

 These fits were obtained for the MEGA galaxies with .

 We have adopted the SDSS10 best-fit model parameters at in order to fit FPZ(); only galaxies with are included in the binned data.

 These are fits of the model to mean M and mean SFR using integrals of double Schechter functions taken from Ilbert et al. (2013), and assuming that sSFR varies with redshift as the star-forming galaxies observed by Karim et al. (2011); there are 28 dof in these fits (6 redshift bins 5 mass bins, since we do not use the lowest-mass bin, and two of the mass bins are empty at the given redshift bin). The redshifts bins and the metallicity behavior with redshift to emulate are taken from mean trends of the MEGA dataset. We have adopted the SDSS10 best-fit model parameters at in order to fit FPZ(). See text for more details.

Table 1: FPZ model best-fit parameters

Despite the differences (O/H calibrations, stellar yields, IMF), our new best-fit FPZ model parameters are generally similar to those of Dayal et al. (2013). The small M dependence of the accretion term, and its small amplitude are consistent with previous values; for the galactic winds, however, we find however a slightly larger coefficient and a slightly steeper M dependence ( vs. ). The inverse SF timescales  are also smaller than in Dayal et al. (2013), corresponding to  Myr vs.  Myr previously. Part of the differences in the best-fit parameters may be due to the choice of the model IMF and our assumptions for the O/H calibration; here we use the Asplund et al. (2009) values of 12log(O/H) = 8.69 for solar oxygen abundance, and a metal mass fraction Z/H of 0.0198222This corresponds to an oxygen abundance of while Dayal et al. (2013) used .. Nevertheless, despite this and the difference in the fitting methods, the results are in reasonably good agreement.

The mass loading factor for galactic winds, , found by the SDSS10 best-fit model is roughly consistent with those found by the cosmological hydrodynamic simulations of Davé et al. (2011), despite their different underlying assumptions about the relation of SFR, gas accretion, and galactic winds. They assume an “equilibrium” model (further described below, Sect. 3.1) in which the infall rate is balanced by the sum of the SFR and the outflow . The results of their simulations show a trend with stellar mass of , or momentum-driven winds, consistent with the MEGA best fit, and only slightly flatter than the power-law index given by the SDSS10 best fit. This power-law M dependence of corresponds roughly to the low-mass linear portion of SDSS10 MZR with a slope of (Paper I).

In contrast, the accretion or infall parameter is significantly different from what is found in some equilibrium models. While the infall amplitude is required to be larger than the outflow in such models, (see, e.g., Davé et al., 2011), we find a much smaller value, roughly a factor of 10 smaller than , and with a much shallower M dependence. These differences will be discussed further in Sect. 3.1. In any case, our model is able to reproduce O/H of the 80 000 SDSS10 galaxies to within  dex, lending credibility to our approach.

The quantity  corresponds approximately to a gas depletion time, with some caveats. Observationally, gas depletion times, , are inferred from the measured ratio of SFR and gas mass,  = /, either locally as surface densities, or globally integrated over the entire galaxy (e.g., Bigiel et al., 2011; Saintonge et al., 2011; Huang & Kauffmann, 2014; Genzel et al., 2015; Hunt et al., 2015). For spiral galaxies, typical depletion times  for the H component are 2.4 Gyr (e.g., Bigiel et al., 2011) but can be as small as 50 Myr in metal-poor starbursts (Hunt et al., 2015) or as large as 10 Gyr in more quiescent systems (e.g., Saintonge et al., 2011; Bothwell et al., 2014). On the other hand, atomic gas Hi depletion times are relatively constant, 3.4 Gyr (Schiminovich et al., 2010). The values we find for  ( Myr for SDSS10,  Myr for MEGA) are shorter than either of these.

The reason for this disagreement can be understood as follows: the constant of proportionality  relating gas mass and SFR (see Eqn. (2)) is in reality the ratio of a (dimensionless) star-formation efficiency  and a SF timescale,  (e.g., Krumholz et al., 2009, 2012), where the subscript “ff” refers to a free-fall or dynamical timescale. Physically,  gives the fraction of gas converted into stars over a dynamical timescale, in a process that is usually highly inefficient with (e.g., Krumholz & Tan, 2007). With such efficiencies and assuming that  = /  the SDSS10 fitted values of  would give  Myr. These times are unrealistically small for typical giant molecular clouds or spiral disks, which instead have typical dynamical times 15-20 Myr (Krumholz et al., 2012).

However, what we call  is in fact only the gas that enriches the ISM with metals as it is recycled through star formation; it is not the total gas available that would be observed. In other words, (model) = (observed) , or from Eqn. (2):


We can use this equation to calculate typical gas depletion times as estimated from our fits of the SDSS10 galaxies. Assuming 0.03, we find 3 Gyr, similar to typical gas depletion times for both the molecular and the atomic component333To explore the molecular gas fraction necessary to form stars would be an important addition to this discussion, but is beyond the scope of this work..

Comparison of the MEGA results with those of SDSS10 shows that the amplitude and power-law index for the wind parameters are consistent with SDSS10, but the accretion parameters and the inverse SF timescales, , differ. As can be seen in Fig. 2, the high-mass end of the SFR dependency of O/H falls off more steeply for the MEGA galaxies than for SDSS10. For the PP04N2 and PP04O3N2 calibrations, gas accretion for the MEGA galaxies seems to increase with decreasing M, in contrast to the behavior of the SDSS10 sample; because of the highly star-forming nature of the MEGA dataset, this could be telling us that the lower-mass galaxies are more gas rich than higher-mass ones. On the other hand, the SF timescales implied by the FPZ model fit to the MEGA galaxies are longer than for SDSS10, although with relatively large uncertainties. More generally, the SDSS10 best-fit parameters are in all cases much better determined than those for the MEGA dataset, probably because of the much larger number of galaxies in the former. Therefore, in what follows, we will adopt the best-fit FPZ parameters for the SDSS10 galaxies to extend the model to high redshift. Table LABEL:tab:models gives the best-fit FPZ model parameters for the SDSS10 and MEGA datasets.

Figure 3: Binned measurements of 12log(O/H) estimated from SFR and the FPZ model at as a function of redshift for the MEGA dataset (left panel) and the simulated dataset (right). In both panels, the symbols correspond to the values of 12log(O/H) that would be inferred using the FPZ() model applied to the median SFR and M within each redshift bin (according to binned M values); the curves show the average behavior of the MEGA dataset as reported in Paper I. See text for a description of the simulated galaxy populations shown in the right panel.

3.1 Comparison with previous work at

As discussed in Paper I, the intention of the FPZ is similar to that of the “Fundamental Metallicity Relation” (“FMR”, Mannucci et al., 2010), namely to relate the metal content of galaxies to their star-formation activity. Nevertheless, the two formulations are different; the planar formulation of the FPZ, accurate to 0.16 dex, extends to galaxies with M as low as  and to redshifts (see Sect. 2 and Paper I). In contrast, the extension to lower stellar masses of the FMR is quadratic (Mannucci et al., 2011), and, as shown in Paper I, has a larger mean dispersion than the FPZ and a significant offset for the MEGA sample which spans a much broader parameter space than the original SDSS dataset at for which the FMR was developed.

Due to its success, which by including SFR reduced significantly the dispersion for 12log(O/H) of the original dataset relative to only stellar mass, many models have focused on reproducing the FMR at by Mannucci et al. (2010). Because of the underlying notion that metallicity is somehow connected to SFR, common to both the FPZ and the FMR formulations, before pursuing our model for the FPZ, here we discuss previous attempts to theoretically understand the FMR.

Despite using simulation runs calibrated to a number observables (stellar mass function, stellar mass-size relation and the stellar mass-black hole relation) the current gold-standard EAGLE simulations are unable to reproduce the local FMR; whilst reproducing observations to within 0.15 dex for high mass (M) galaxies, they severely over-estimate the metal content of lower mass halos by as much as 0.4 dex (Lagos et al., 2016). Given the sub-grid limitations of simulation, most effort has thus been diverted to developing an analytic understanding of the FMR.

We briefly discuss the two previous works most close in spirit to ours. The Davé et al. (2012) model assumes every galaxy to be in “equilibrium” with the infall rate exactly balancing the gas lost in outflows and star formation. While this model correctly yields a metallicity that is independent of star formation for high-mass galaxies, it fails to capture the metallicity downturn observed at low mass. Interestingly, however, this model finds outflows to be momentum driven with a mass power-law dependence of , which however we obtain only for the MEGA sample; the SDSS10 slopes are slightly steeper than this, . Peeples & Shankar (2011) find that reproducing the local mass-metallicity relation requires a high metal-expulsion efficiency (compared to expelled gas, “preferentially metal-enriched winds”) that scales steeply with increasing M. Although this might yield the observed FMR (or FPZ) for low-mass galaxies, it is doubtful if such an assumption could reproduce the constant metallicity observed for the most massive systems, independently of SFRs ranging over five orders of magnitude presented in the MEGA sample.

Lilly et al. (2013) use the “gas-regulator” model wherein stars are embedded in a gas reservoir whose mass is increased by infall, and the SFR being proportional to the gas mass available at any time. Assuming both the star formation efficiency and outflow rate to be independent of the halo mass (the “ideal regulator”), these authors find the metallicity to be largely independent of the evolutionary path because of gas continually “flushing” the system. This is quite contrary to our model that requires outflows to scale with the halo mass, and finds the metallicity to explicitly depend on the gas evolution history.

Figure 4: Binned measurements of 12log(O/H) estimated from SFR and the FPZ() model with () added to the FMZ() model as a function of redshift for the MEGA dataset (left panel) and the simulated dataset (right). In both panels, the symbols correspond to the (PP04N2) values of 12log(O/H) that would be inferred using the FPZ() model applied to the median SFR and M within each redshift bin (according to binned M values); the curves show the average behavior of the MEGA dataset as reported in Paper I. See text for a description of the simulated galaxy populations shown in the right panel.

Figure 5: Binned measurements of 12log(O/H) estimated from SFR and the FPZ() model with , added to the FMZ() model as a function of redshift for the MEGA dataset (left panel) and the simulated dataset (right). In both panels, the symbols correspond to the (PP04N2) values of 12log(O/H) that would be inferred using the FPZ() model applied to the median SFR and M within each redshift bin (according to binned M values); the curves show the average behavior of the MEGA dataset as reported in Paper I. See text for a description of the simulated galaxy populations shown in the right panel.

Figure 6: Binned measurements of 12log(O/H) estimated from SFR and the FPZ() model with () added to the FMZ() model as a function of redshift for the MEGA dataset (left panel) and the simulated dataset (right). In both panels, the symbols correspond to the (PP04N2) values of 12log(O/H) that would be inferred using the FPZ() model applied to the median SFR and M within each redshift bin (according to binned M values); the curves show the average behavior of the MEGA dataset as reported in Paper I. See text for a description of the simulated galaxy populations shown in the right panel.

3.2 Testing the model at higher redshift

In order to extend the FPZ model to , we first test whether the SDSS10 best-fit parameters apply to the data at higher redshift. Such behavior would be expected if the lower metallicities at high redshift were compensated for merely by the more intense star-formation activity, which could power larger outflow of metals through stronger galactic winds. The left panel of Fig. 3 shows the behavior of the MEGA sample if we infer metallicity through the FPZ() model parameters. The curves are the fits to the observed O/H of the MEGA dataset as a function of redshift as reported in Paper I; the symbols correspond to the values of 12log(O/H) that would be inferred from the FPZ model applied to the median SFR and M in each redshift bin. In all M bins, up to the FPZ correctly predicts the observed O/H. However, for higher redshifts, given the observed (median) SFRs, the FPZ model overpredicts the observed O/H, in particular for the highest M mass bins where the discrepancy is 0.4 dex. On the other hand, in the lowest mass bin (log(M/) =8.5-9.0) O/H is underestimated by 0.1 dex. Similar tensions relative to the model by Dayal et al. (2013) are seen also in the sample of galaxies by Grasshorn Gebhardt et al. (2016) where they find that galaxies with M have higher O/H relative to their local counterparts, while lower-mass systems have lower abundances.

Because the high SFRs of the MEGA dataset may not be fully representative of typical high- “main-sequence” galaxy populations (see Paper I), as a check, we have simulated truly main-sequence galaxy populations as a function of redshift using the Galaxy Stellar Mass Functions (GSMFs) by Ilbert et al. (2013) for and for the GSMF from Baldry et al. (2012). For a given M, the SFR has been inferred by using the SFMS formulation of Speagle et al. (2014) as a function of cosmic time. For each redshift bin, integration over the appropriate weighted GSMFs was performed to obtain the mean M and the mean SFR at that redshift for main-sequence populations. Using these values of M and SFR, we then calculated O/H using the FPZ() model parameters as for the MEGA data. The result is shown in the right panel of Fig. 3, where the metallicities predicted from the FPZ() model are compared to the mean observed O/H behavior (shown by curves) of the MEGA dataset as reported in Paper I. As in the left panel, symbols correspond to the metallicities that would be inferred for the simulated galaxies of a given M and SFR(M,) from the FPZ() model within each M and redshift bin. Despite the two very different approaches, both the simulated galaxies (right panel of Fig. 3) and the MEGA galaxies (left panel) show similar behavior. The FPZ() parameters are successful to , but fail in the same way as for the MEGA dataset at higher . The consistency of the simulations and the MEGA galaxies is encouraging, and in the next section we use both methods to extend the FPZ model to high redshift.

4 Extending the model to higher redshift

As illustrated in the previous section, the higher SFRs at are insufficient by themselves through the FPZ() model to lower the metallicities to the levels observed. We have thus investigated three avenues of adapting the FPZ model to : (a) changes in , since timescales and/or star-formation efficiencies might be expected to change with redshift (Sect. 4.1); (b) redshift variations of accretion and wind parameters ( and ) (Sect. 4.2); and (c) higher gas fractions through possible changes in the parameter ( Sect. 4.3, see Eqn. (5)). Since the discrepancies in Fig. 3 seem to depend on M, we considered separate dependencies on M and redshift (or cosmic time). Thus, for each approach to establish the -dependent FPZ model, we fix FPZ() to the SDSS10 parameters and introduce a scaling factor for either , (, ), or . Thus only two parameters are to be fit: one for a possible M scaling for (, ) and one for a scaling with redshift (, , , ). Although these are only a few of the many possible adaptations of the FPZ model to high , we have limited the possibilities for simplicity.

To constrain the model, we have assumed that the true metallicities of high- galaxy populations are approximated by the mean behavior of the MEGA dataset as reported in Paper I. The best-fitting model parameters for are obtained by minimizing the residuals of the observed mean behavior of 12log(O/H) relative to the model over the various mass and redshift bins, not including ; because of small numbers at high redshift, we also do not consider the lowest mass bin (log(M/)8.5). The resulting degrees of freedom in the fits are 25 for the MEGA dataset (since some M bins are missing at ), and 28 for the simulated galaxies (5 M bins over 6 redshift bins, 2 free parameters). As mentioned above, in all high- FPZ models, we adopted the SDSS10 parameters and adjusted  (first approach), and , the accretion and wind coefficients (second), or (third). Table LABEL:tab:models gives the best-fit FPZ() (D02, PP04N2, PP04O3N2) model parameters for the three approaches which are described in the following sections.

4.1 Redshift variation of star-formation timescales and/or efficiencies

We first investigated whether scaling  with M and could have the desired effect of lowering O/H at high redshift. Several methods were explored for introducing a scaling factor  applied as a multiplicative factor to  in the FPZ models. Although we studied numerous ways to scale , including a mass-quenching scenario as formulated by Peng et al. (2010), only two gave reasonably low mean O/H residuals,  dex for both MEGA and simulated galaxies. These correspond to () = (0) where  is given by:

  1. (M/ 

  2. (M/ 

where (0) gives the value of  at , age and age are the ages of the universe at redshift and , respectively, and and are the parameters to be fit. Since the two fits are equivalent in terms of quality (low residuals), we chose (a) since it is generally easier to formulate numerically.

The values of 12log(O/H) predicted by this model as a function of redshift are shown in Fig. 4; as in Fig. 3 the left panel gives the MEGA dataset and the right panel the simulated galaxies. The fit is excellent for the simulated galaxies, with a mean residual  dex (except for D02), but for the MEGA dataset is slightly worse,  dex. There are some problems in the intermediate- regime where the metallicities inferred from the FPZ model by varying  with mass and tend to be overestimated. The predictions for low-mass galaxies also tend to be overestimated at low redshift.

4.2 Redshift variation of gas accretion and galactic winds

The second avenue of investigation relied on simultaneously scaling the gas accretion coefficient and the galactic wind coefficient , in order to quantify whether the scaling with SFR changes with redshift. For simplicity, we used one coefficient to define the redshift variation of , and another for that of :

where and are the values of and at . The implicit assumption is that the M dependence of and given by the SDSS10 fit in Table LABEL:tab:models holds also at .

Figure 6 gives the results of this fit, where, as before, the left panel shows the MEGA dataset and the right the simulated galaxies. The fit for the MEGA dataset is good, with a mean residual between data and predicted O/H of  dex, similar to the previous  approach described above. For the simulated galaxies, the fit is still good ( dex), although slightly worse than the  model (except for D02 where it is better).

4.3 Redshift variation of gas fraction

The last successful formulation involved scaling the model parameter which basically defines the gas-mass fraction relative to its initial value in the galaxy’s evolution (see Eqns. (5) and (6)). For computational reasons, the scaling factor was introduced in the denominator of :

The FPZ() model predictions of O/H with this approach are shown in Fig. 6 where the MEGA dataset is shown in the left panel and the simulated galaxies in the right. Again, the fit for the MEGA dataset is good, with mean residuals  dex, consistent with the previous two formulations. The behavior of the MEGA galaxies is similar to the  approach described in Sect. 4.1 where O/H is over-predicted by the model in the intermediate- redshift range, although the discrepancies at the low-mass end are slightly less pronounced. On the other hand, this formulation of the FPZ() model for the simulated populations is excellent, with a mean residual of  dex.

5 Discussion

This is one of the first works that aims to study the physics governing the redshift-evolution of the FPZ (or FMR). Lilly et al. (2013) predicted an un-evolving FPZ unless the parameters governing inflow/outflow/star formation evolve with redshift; our approach suggests that indeed these parameters evolve. Yates et al. (2012) study the evolution of the MZR from to by comparing the semi-analytic models of Guo et al. (2011) with an SDSS-selected sample similar to SDSS10. However, their models were unable to predict the observed decrease of metallicity for a fixed M, and they concluded that “chemical enrichment in the model galaxies proceeded too rapidly at early times”.

Here we have shown that to correctly reproduce the coevolution of SFR and O/H, i.e., correctly fit the observations, that the model describing the co-dependence of M, SFR, and metallicity must change with redshift. All three FPZ() fits to both the MEGA dataset and the simulated galaxies as described above successfully predict the O/H trends with redshift to within  dex. Although the three formulations are apparently describing different physical manifestations of the coevolution of metallicity and SFR, they are generally painting a similar picture, namely an increase of gas content with redshift.

5.1 Redshift variation of the model

According to Table LABEL:tab:models, the scaling of the FPZ() model with  suggests that:


The above (), () coefficients are approximate, taken as a rough average of the (PP04N2) simulated and MEGA datasets. Eqn. (8) would imply that at a given M, , roughly the product of gas depletion time  and SF efficiency , is increasing slightly with redshift (roughly 30% larger at ). Assuming that  remains constant, this would be contrary to observational evidence that suggests that molecular depletion times are shorter at higher redshift (e.g., Genzel et al., 2010, 2015; Silverman et al., 2015). Galaxies are also more gas-rich at higher redshift, with as much as 4060% of their dynamical mass in H (e.g., Tacconi et al., 2010; Daddi et al., 2010), a fraction 510 times higher than typical spiral disks at . However, as discussed in Sect. 3, depletion times are defined observationally as the ratio of observed gas mass to SFR. Our models do not know about observed gas mass; they only consider the gas that is enriching the ISM with metals, i.e., the gas that is effectively converted into stars. Thus, of necessity, our models cannot distinguish between the observationally defined  and the product of  and , or equivalently the ratio of  and . Given that  = (observed) / (see Eqn. (7)) it is likely that the  FPZ model scaling is pointing to increasing gas content with , rather than longer depletion times.

The FPZ() formulation with the accretion and wind coefficients is consistent with this. As shown in Table LABEL:tab:models, the scaling of the model with and :


indicates that the mass loading relative to SFR in the galactic outflows powered by star formation is roughly the same at high as at . In contrast, the mass loading in accretion is significantly increased, almost 3 times higher at than locally. Thus, to explain the drop in metallicity at high redshift, more (pristine) gas is needed, in the model acquired through accretion. Our result is roughly consistent with Papovich et al. (2011) who find that gas fractions increase roughly as from to , and after this “accretion epoch”, at lower redshifts both accretion and SFR are reduced.

A similar indication is given by the FPZ() model based on the redshift variation of , equal to the ratio of  and the initial gas mass (see Eqn. (5)). The coefficients in Table LABEL:tab:models indicate that:


Both the M and redshift dependencies are similar to (although slightly smaller than) those for the  formulation; this is not unexpected given that . The approach is perhaps more direct, but is essentially indicating, as above, that more gas is needed with increasing redshift, in order to form stars at the necessary levels, recycle the ISM, and achieve the observed reduced metallicities.

The increase in gas content with redshift roughly given by Eqns. (8) and (10) is similar to (the inverse of) that found observationally for molecular depletion times  of main-sequence galaxies by Genzel et al. (2015). Since, as discussed above, our models are unable to separate gas content from  and , this is an encouraging consistency. Moreover, for massive galaxies with M, we would predict gas fractions 7 times higher at than at , roughly consistent with the CO observations by Geach et al. (2011), and only slightly lower then the increase of a factor of 10 found by Bothwell et al. (2013b) for luminous sub-millimeter galaxies.

Nevertheless, a limitation of our model is that it does not explicitly distinguish between accreted gas, presumably Hi, and the gas which forms stars that, at these high SFRs, must be molecular. Indeed, the changes with redshift of relative gas content predicted by the FPZ() model are slightly lower than the results derived observationally for H; Tacconi et al. (2013) and Saintonge et al. (2013) find that, as goes from 2 to 1, gas fractions decrease by a factor of 1.4, while we would predict a change of only 1.1-1.2 (for a galaxy of fixed stellar mass). The gas content of our model comprises not only the molecular component but also Hi; although it is not yet possible to observe Hi at high , it is likely that Hi content shows a smaller redshift variation than H (e.g., Lagos et al., 2011; Popping et al., 2012; Lagos et al., 2016). Thus, total gas fractions are expected to increase less rapidly with redshift than H alone. More detailed comparisons with observations will require observations of atomic gas to cosmological redshifts which should be possible with the Square Kilometer Array (SKA).

5.2 Redshift variation of the stellar mass scaling

It is well established that at a given redshift the gas fraction in massive galaxies is lower than in less massive ones (e.g., Saintonge et al., 2011; Popping et al., 2012; Huang et al., 2012; Boselli et al., 2014; Bothwell et al., 2014; Popping et al., 2015b). Moreover, as discussed above, it is clear that gas fraction increases with redshift. The FPZ() model with the  and approaches also predicts that the amount of increase in gas content with redshift varies with stellar mass. In particular, the M scaling of the FPZ() models (see Eqns. (8) and (10)) suggests that the gas content of high-mass galaxies relative to less massive ones increases with increasing redshift. For a galaxy with M =  at , the growth in gas content is predicted to be times that at ; a more massive galaxy, M =  at the same redshift the increase would be times the gas content of a galaxy of the same mass at . A less massive galaxy, with M = , would be expected to have roughly the same gas content at as at . Such a result would be consistent with “downsizing” in which massive galaxies evolve more rapidly than lower-mass ones, thus consuming their gas at earlier epochs (e.g., Cowie et al., 1996; De Lucia et al., 2006; Bundy et al., 2006; Thomas et al., 2010). Specifically, Thomas et al. (2010) found that SF activity in galaxies with M  peaked at increasingly higher redshifts, , well within the redshift and mass ranges probed by the MEGA and simulated galaxies.

Such a trend of gas fraction with stellar mass and redshift is also in accord with some observational estimates of the mass variations of gas fraction with redshift, either through indirect determinations of gas content by inverting gas-SFR scaling relations (e.g., Popping et al., 2012, 2015b), or with gas masses inferred from measured dust masses up to after correcting for metallicity (Santini et al., 2014). Santini et al. (2014) find that from to , the gas fraction in massive galaxies decreases more sharply than in lower-mass ones; less massive galaxies show a shallower decrease in the gas content.

However, observations of H alone are in disagreement with this. Morokuma-Matsui & Baba (2015) estimated redshift variations of H fraction through a compilation of CO observations and find that more massive galaxies show less evolution in H fraction than lower-mass ones, in direct contrast to our results. The same contradiction emerges in the study by Dessauges-Zavadsky et al. (2015) who studied five lensed galaxies and compared them with CO observations from the literature. Genel et al. (2014) Illustris simulations also find that less massive galaxies at higher redshift have a larger change in gas fraction with redshift. Observationally, our result for galaxies with masses  is very difficult to verify given that virtually all observational studies so far of gas content at high redshift are limited to M (although see Saintonge et al., 2013; Dessauges-Zavadsky et al., 2015, who studied lensed galaxies). Nevertheless, the results of the FPZ() models suggest that massive galaxies at relative to must have a higher gas content in order to achieve the relatively lower metallicities. Incorporating a multi-phase approach as in Magrini et al. (2012) might help interpret the different gas components in our models.

One limitation of the FPZ model was discussed above, namely the inability to distinguish between H, Hi, and total gas content. At least one other possible limitation to the model is the lack of mass scaling of , or equivalently the SF timescale. Such as omission can be justified for the SDSS10 galaxies because, as discussed in Sect. 3, in typical spiral disks both molecular depletion times  and Hi depletion time are roughly constant,  Gyr (Bigiel et al., 2011; Catinella et al., 2010; Schiminovich et al., 2010). Nevertheless, this could be a problem for the MEGA dataset because of its highly star-forming nature, and large variation in M. It could also be troublesome for the high- main-sequence simulated galaxies because of their relatively higher SFRs compared to local ones. It is possible that the M scaling we find for the  and adaptations of the FPZ model to is a consequence of neglecting such a treatment at .

Our treatment of trends of FPZ parameters with and M as simple power laws is also highly over-simplified. As found by Lagos et al. (2011); Lagos et al. (2014); Morokuma-Matsui & Baba (2015), the redshift and M dependence of gas-mass fractions is more complicated than this, as there are inflections and slight curvatures in the behavior of both quantities. Thus, our models are almost certainly not exact, but rather a simplified scaling aimed at a general representation of the trends necessary to explain the coevolution of SFR and metallicity.

Recently it has been proposed that scaling laws based on gas fraction, SFR, and M are more fundamental than those based on metallicity (e.g., Bothwell et al., 2013a; Santini et al., 2014; Lagos et al., 2016; Bothwell et al., 2016). The physical basis of such a scaling is clear, since gas provides the fuel for star formation and the reservoir of gas-phase metals. However, the role of Hi vs. H is still under debate, and the possibility of verifying through observations the amount of total gas in Hi is currently not possible, although SKA will be able to help shed light on this problem.

6 Summary

In a companion paper, we constructed a new MEGA dataset for studying the coevolution of metallicity and SFR; the data are compiled from 19 different samples up to , spanning a factor of in M, in SFR, and almost two orders of magnitude in O/H. As a comparison sample, we include the SDSS10 galaxies studied by Mannucci et al. (2010).

  Here we update the model of Dayal et al. (2013) for the behavior of M, SFR, and O/H and apply it to the SDSS10 and MEGA datasets; the new model relies on the stellar yields from Vincenzo et al. (2016) and Nomoto et al. (2013) assuming a Chabrier (2003) IMF. With only 5 free parameters, the FPZ model at is able to reproduce the observed oxygen abundance of the 80 000 SDSS10 galaxies to within 0.050.06 dex, and of the MEGA galaxies to within  dex.

  We have extended the FPZ() model to higher redshift by exploring three possibilities: (1) a redshift (and mass) scaling of SF timescale  (Sect. 4.1); (2) a redshift scaling of accretion and galactic wind mass loading (Sect. 4.2); and (3) a redshift (and mass) scaling of gas fraction (Sect. 4.3). These approaches give similar results, and are able to reproduce the observed metallicity trends in the MEGA sample to within  dex.

  The extension of the FPZ() model allows us to quantify how gas accretion and outflow depend on redshift. Although the specific mass loading of outflows does not change measurably during the evolution, the accretion rate and gas content of galaxies increase significantly with redshift. These two effects explain, either separately or possibly in tandem, the observed lower metal abundance of high- galaxies.


We thank Filippo Fraternali for interesting discussion and insightful comments. This work has benefited from the DAVID network ( for fostering a fruitful collaborative environment. PD gladly acknowledges funding from the EU COFUND Rosalind Franklin program, and LH from a national research grant, PRIN-INAF/2012.


  • Andrews & Martini (2013) Andrews, B. H., & Martini, P. 2013, ApJ, 765, 140
  • Asplund et al. (2009) Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481
  • Baldry et al. (2012) Baldry, I. K., Driver, S. P., Loveday, J., et al. 2012, MNRAS, 421, 621
  • Bigiel et al. (2011) Bigiel, F., Leroy, A. K., Walter, F., et al. 2011, ApJ, 730, L13
  • Boselli et al. (2014) Boselli, A., Cortese, L., Boquien, M., et al. 2014, A&A, 564, A66
  • Bothwell et al. (2016) Bothwell, M. S., Maiolino, R., Peng, Y., et al. 2016, MNRAS, 455, 1156
  • Bothwell et al. (2013a) Bothwell, M. S., Maiolino, R., Kennicutt, R., et al. 2013b, MNRAS, 433, 1425
  • Bothwell et al. (2013b) Bothwell, M. S., Smail, I., Chapman, S. C., et al. 2013a, MNRAS, 429, 3047
  • Bothwell et al. (2014) Bothwell, M. S., Wagg, J., Cicone, C., et al. 2014, MNRAS, 445, 2599
  • Brinchmann et al. (2004) Brinchmann, J., Charlot, S., White, S. D. M., et al. 2004, MNRAS, 351, 1151
  • Bundy et al. (2006) Bundy, K., Ellis, R. S., Conselice, C. J., et al. 2006, ApJ, 651, 120
  • Catinella et al. (2010) Catinella, B., Schiminovich, D., Kauffmann, G., et al. 2010, MNRAS, 403, 683
  • Chabrier (2003) Chabrier, G. 2003, PASP, 115, 763
  • Cowie et al. (1996) Cowie, L. L., Songaila, A., Hu, E. M., & Cohen, J. G. 1996, AJ, 112, 839
  • Cresci et al. (2012) Cresci, G., Mannucci, F., Sommariva, V., et al. 2012, MNRAS, 421, 262
  • Cullen et al. (2014) Cullen, F., Cirasuolo, M., McLure, R. J., Dunlop, J. S., & Bowler, R. A. A. 2014, MNRAS, 440, 2300
  • Daddi et al. (2010) Daddi, E., Bournaud, F., Walter, F., et al. 2010, ApJ, 713, 686
  • Davé et al. (2011) Davé, R., Finlator, K., & Oppenheimer, B. D. 2011, MNRAS, 416, 1354
  • Davé et al. (2012) Davé, R., Finlator, K., & Oppenheimer, B. D. 2012, MNRAS, 421, 98
  • Dayal et al. (2013) Dayal, P., Ferrara, A., & Dunlop, J. S. 2013, MNRAS, 430, 2891
  • de los Reyes et al. (2015) de los Reyes, M. A., Ly, C., Lee, J. C., et al. 2015, AJ, 149, 79
  • De Lucia et al. (2006) De Lucia, G., Springel, V., White, S. D. M., Croton, D., & Kauffmann, G. 2006, MNRAS, 366, 499
  • Denicoló et al. (2002) Denicoló, G., Terlevich, R., & Terlevich, E. 2002, MNRAS, 330, 69 (D02)
  • Dessauges-Zavadsky et al. (2015) Dessauges-Zavadsky, M., Zamojski, M., Schaerer, D., et al. 2015, A&A, 577, A50
  • Elbaz et al. (2011) Elbaz, D., Dickinson, M., Hwang, H. S., et al. 2011, A&A, 533, A119
  • Erb et al. (2006a) Erb, D. K., Shapley, A. E., Pettini, M., et al. 2006a, ApJ, 644, 813
  • Eyles et al. (2007) Eyles, L. P., Bunker, A. J., Ellis, R. S., et al. 2007, MNRAS, 374, 910
  • Fraternali & Binney (2008) Fraternali, F., & Binney, J. J. 2008, MNRAS, 386, 935
  • Fraternali & Tomassetti (2012) Fraternali, F., & Tomassetti, M. 2012, MNRAS, 426, 2166
  • Gavazzi et al. (2015) Gavazzi, G., Consolandi, G., Dotti, M., et al. 2015, A&A, 580, A116
  • Geach et al. (2011) Geach, J. E., Smail, I., Moran, S. M., et al. 2011, ApJ, 730, L19
  • Genzel et al. (2010) Genzel, R., Tacconi, L. J., Gracia-Carpio, J., et al. 2010, MNRAS, 407, 2091
  • Genzel et al. (2015) Genzel, R., Tacconi, L. J., Lutz, D., et al. 2015, ApJ, 800, 20
  • Genel et al. (2014) Genel, S., Vogelsberger, M., Springel, V., et al. 2014, MNRAS, 445, 175
  • Grasshorn Gebhardt et al. (2016) Grasshorn Gebhardt, H. S., Zeimann, G. R., Ciardullo, R., et al. 2016, ApJ, 817, 10
  • Guo et al. (2011) Guo, Q., White, S., Boylan-Kolchin, M., et al. 2011, MNRAS, 413, 101
  • Henry et al. (2013) Henry, A., Martin, C. L., Finlator, K., & Dressler, A. 2013, ApJ, 769, 148
  • Hopkins et al. (2008) Hopkins, A. M., McClure-Griffiths, N. M., & Gaensler, B. M. 2008, ApJ, 682, L13
  • Hopkins et al. (2012) Hopkins, P. F., Quataert, E., & Murray, N. 2012, MNRAS, 421, 3522
  • Huang & Kauffmann (2014) Huang, M.-L., & Kauffmann, G. 2014, MNRAS, 443, 1329
  • Huang et al. (2012) Huang, S., Haynes, M. P., Giovanelli, R., & Brinchmann, J. 2012, ApJ, 756, 113
  • Hunt et al. (2015) Hunt, L. K., García-Burillo, S., Casasola, V., et al. 2015, A&A, 583, A114
  • Hunt et al. (2012) Hunt, L., Magrini, L., Galli, D., et al. 2012, MNRAS, 427, 906
  • Hunt et al. (2016) Hunt, L., Dayal, P., Magrini, L., & Ferrara, A.  2016, MNRAS, submitted
  • Ilbert et al. (2013) Ilbert, O., McCracken, H. J., Le Fèvre, O., et al. 2013, A&A, 556, A55
  • Karim et al. (2011) Karim, A., Schinnerer, E., Martínez-Sansigre, A., et al. 2011, ApJ, 730, 61
  • Kewley & Dopita (2002) Kewley, L. J., & Dopita, M. A. 2002, ApJS, 142, 35 (KD02)
  • Kewley & Ellison (2008) Kewley, L. J., & Ellison, S. L. 2008, ApJ, 681, 1183
  • Krumholz & Dekel (2012) Krumholz, M. R., & Dekel, A. 2012, ApJ, 753, 16
  • Krumholz et al. (2012) Krumholz, M. R., Dekel, A., & McKee, C. F. 2012b, ApJ, 745, 69
  • Krumholz et al. (2009) Krumholz, M. R., McKee, C. F., & Tumlinson, J. 2009, ApJ, 699, 850
  • Krumholz & Tan (2007) Krumholz, M. R., & Tan, J. C. 2007, ApJ, 654, 304
  • Lagos et al. (2011) Lagos, C. D. P., Baugh, C. M., Lacey, C. G., et al. 2011, MNRAS, 418, 1649
  • Lagos et al. (2014) Lagos, P., Papaderos, P., Gomes, J. M., Smith Castelli, A. V., & Vega, L. R. 2014, A&A, 569, A110
  • Lagos et al. (2016) Lagos, C. d. P., Theuns, T., Schaye, J., et al. 2016, MNRAS, 459, 2632
  • Lee et al. (2015) Lee, N., Sanders, D. B., Casey, C. M., et al. 2015, ApJ, 801, 80
  • Lilly et al. (2013) Lilly, S. J., Carollo, C. M., Pipino, A., Renzini, A., & Peng, Y. 2013, ApJ, 772, 119
  • Ly et al. (2015) Ly, C., Rigby, J. R., Cooper, M., & Yan, R. 2015, ApJ, 805, 45
  • Magrini et al. (2012) Magrini, L., Hunt, L., Galli, D., et al. 2012, MNRAS, 427, 1075
  • Maiolino et al. (2008) Maiolino, R., Nagao, T., Grazian, A., et al. 2008, A&A, 488, 463
  • Mannucci et al. (2009) Mannucci, F., Cresci, G., Maiolino, R., et al. 2009, MNRAS, 398, 1915
  • Mannucci et al. (2010) Mannucci, F., Cresci, G., Maiolino, R., Marconi, A., & Gnerucci, A. 2010, MNRAS, 408, 2115
  • Mannucci et al. (2011) Mannucci, F., Salvaterra, R., & Campisi, M. A. 2011, MNRAS, 414, 1263
  • Martin (2005) Martin, C. L. 2005, ApJ, 621, 227
  • Morokuma-Matsui & Baba (2015) Morokuma-Matsui, K., & Baba, J. 2015, MNRAS, 454, 3792
  • Noeske et al. (2007) Noeske, K. G., Weiner, B. J., Faber, S. M., et al. 2007, ApJ, 660, L43
  • Nomoto et al. (2013) Nomoto, K., Kobayashi, C., & Tominaga, N. 2013, ARA&A, 51, 457
  • Onodera et al. (2016) Onodera, M., Carollo, C. M., Lilly, S., et al. 2016, ApJ, 822, 42
  • Pacifici et al. (2015) Pacifici, C., da Cunha, E., Charlot, S., et al. 2015, MNRAS, 447, 786
  • Papovich et al. (2011) Papovich, C., Finkelstein, S. L., Ferguson, H. C., Lotz, J. M., & Giavalisco, M. 2011, MNRAS, 412, 1123
  • Peeples & Shankar (2011) Peeples, M. S., & Shankar, F. 2011, MNRAS, 417, 2962
  • Peng et al. (2010) Peng, Y.-j., Lilly, S. J., Kovač, K., et al. 2010, ApJ, 721, 193
  • Pettini & Pagel (2004) Pettini, M., & Pagel, B. E. J. 2004, MNRAS, 348, L59 (PP04)
  • Popping et al. (2012) Popping, G., Caputi, K. I., Somerville, R. S., & Trager, S. C. 2012, MNRAS, 425, 2386
  • Popping et al. (2015a) Popping, G., Caputi, K. I., Somerville, R. S., & Trager, S. C. 2015a, MNRAS, 452, 169
  • Popping et al. (2015b) Popping, G., Caputi, K. I., Trager, S. C., et al. 2015b, MNRAS, 454, 2258
  • Romano et al. (2010) Romano, D., Karakas, A. I., Tosi, M., & Matteucci, F. 2010, A&A, 522, A32
  • Saintonge et al. (2011) Saintonge, A., Kauffmann, G., Wang, J., et al. 2011, MNRAS, 415, 61
  • Saintonge et al. (2013) Saintonge, A., Lutz, D., Genzel, R., et al. 2013, ApJ, 778, 2
  • Salim et al. (2007) Salim, S., Rich, R. M., Charlot, S., et al. 2007, ApJS, 173, 267
  • Salmon et al. (2015) Salmon, B., Papovich, C., Finkelstein, S. L., et al. 2015, ApJ, 799, 183
  • Santini et al. (2014) Santini, P., Maiolino, R., Magnelli, B., et al. 2014, A&A, 562, A30
  • Schaye et al. (2015) Schaye, J., Crain, R. A., Bower, R. G., et al. 2015, MNRAS, 446, 521
  • Schiminovich et al. (2010) Schiminovich, D., Catinella, B., Kauffmann, G., et al. 2010, MNRAS, 408, 919
  • Silverman et al. (2015) Silverman, J. D., Daddi, E., Rodighiero, G., et al. 2015, ApJ, 812, L23
  • Speagle et al. (2014) Speagle, J. S., Steinhardt, C. L., Capak, P. L., & Silverman, J. D. 2014, ApJS, 214, 15
  • Stark et al. (2009) Stark, D. P., Ellis, R. S., Bunker, A., et al. 2009, ApJ, 697, 1493
  • Steidel et al. (2014) Steidel, C. C., Rudie, G. C., Strom, A. L., et al. 2014, ApJ, 795, 165
  • Tacconi et al. (2010) Tacconi, L. J., Genzel, R., Neri, R., et al. 2010, Nature, 463, 781
  • Tacconi et al. (2013) Tacconi, L. J., Neri, R., Genzel, R., et al. 2013, ApJ, 768, 74
  • Thomas et al. (2010) Thomas, D., Maraston, C., Schawinski, K., Sarzi, M., & Silk, J. 2010, MNRAS, 404, 1775
  • Tremonti et al. (2004) Tremonti, C. A., Heckman, T. M., Kauffmann, G., et al. 2004, ApJ, 613, 898 (T04)
  • Troncoso et al. (2014) Troncoso, P., Maiolino, R., Sommariva, V., et al. 2014, A&A, 563, AA58
  • Veilleux et al. (2005) Veilleux, S., Cecil, G., & Bland-Hawthorn, J. 2005, ARA&A, 43, 769
  • Vincenzo et al. (2016) Vincenzo, F., Matteucci, F., Belfiore, F., & Maiolino, R. 2016, MNRAS, 455, 4183
  • Whitaker et al. (2014) Whitaker, K. E., Franx, M., Leja, J., et al. 2014, ApJ, 795, 104
  • Wuyts et al. (2011) Wuyts, S., Förster Schreiber, N. M., van der Wel, A., et al. 2011, ApJ, 742, 96
  • Wuyts et al. (2014) Wuyts, E., Kurk, J., Förster Schreiber, N. M., et al. 2014, ApJ, 789, L40
  • Wyder et al. (2007) Wyder, T. K., Martin, D. C., Schiminovich, D., et al. 2007, ApJS, 173, 293
  • Xia et al. (2012) Xia, L., Malhotra, S., Rhoads, J., et al. 2012, AJ, 144, 28
  • Yates et al. (2012) Yates, R. M., Kauffmann, G., & Guo, Q. 2012, MNRAS, 2572
  • Yabe et al. (2012) Yabe, K., Ohta, K., Iwamuro, F., et al. 2012, PASJ, 64, 60
  • Zahid et al. (2014) Zahid, H. J., Kashino, D., Silverman, J. D., et al. 2014, ApJ, 792, 75
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
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

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 description