Filename
stringlengths 22
64
| Paragraph
stringlengths 8
5.57k
|
|---|---|
Processed_Chemistry_in_Infrared_Dark_Cloud_Clumps:_a_Molecul.txt | Optical depth calculations show that the N2H+ line is mostly optically thin (median of 0.8) and the C2H line is moderately optically thick (median of 2.4). HCO+ and HNC lines are optically thick (medians of 21 and 19, re- spectively), while their isotopologues are optically thin (median of 0.4 for both). N2H+ opacities show no vari- ations with the evolution of the clumps, whereas C2H, HCO+ and HNC show a slight decrease with the rise of star formation activity. |
Processed_Chemistry_in_Infrared_Dark_Cloud_Clumps:_a_Molecul.txt | In general, column densities of the different molecules change for the different evolutionary stages defined by Chambers et al. (2009) and increase with the evolution of the clumps, with the exception of C2H. However, this is not generally true for molecular abundances (i.e., af- ter dividing by the total H2 column density inferred from 1.2 mm continuum emission). Only the increases of N2H+ and HCO+ abundances are statistically signif- icant and reflect chemical evolution. This is consistent with the results of Busquet et al. (2011), who included both molecules in their chemical modeling of a massive star-forming region. Although it is expected a rise of the HCO+ abundance with the evolution of the clumps, it is not clear why N2H+ also follows this trend. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | ABSTRACT We explore the constraints that can be placed on the evolutionary timescales for typical low redshift galaxies evolving from the blue cloud through the green valley and onto the red se- quence. We utilise galaxies from the GAMA survey with 0.1 < z < 0.2 and classify them according to the intrinsic (u∗ − r∗) colours of their stellar populations, as determined by fits to their multi-wavelength spectral energy distributions. Using these fits to also determine stellar population ages and star formation timescales, we argue that our results are consistent with a green valley population dominated by galaxies that are simply decreasing their star formation (running out of gas) over a timescale of 2-4 Gyr which are seen at a specific epoch in their evolution (approximately 1.6 e-folding times after their peak in star formation). If their fitted star formation histories are extrapolated forwards, the green galaxies will further redden over time, until they attain the colours of a passive population. In this picture, no specific quench- ing event which cuts-off their star formation is required, though it remains possible that the decline in star formation in green galaxies may be expedited by internal or external forces. However, there is no evidence that green galaxies have recently changed their star formation timescales relative to their previous longer term star formation histories. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | forming to passively evolving, as in e.g. Fang et al. (2013) and the review by Salim (2014). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Further, Bell et al. (2003) observed that at z < 1 stars form predominantly in blue galaxies, but stellar mass accumulates in red galaxies, a point recently emphasised by Eales et al. (2015). This demonstrates the importance of galaxies transitioning through the green valley, from blue to red, to our understanding of the last ∼ 7 Gyr of galaxy evolution in general. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | with intermediate levels of star formation must in general have green colours at some point while transitioning towards being ‘red and dead’ (Bremer et al. 2018, henceforth Paper I, and references therein). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | To try to answer these questions and explore potential transi- tion populations in general, as in Paper I we here take advantage of the rich panchromatic Galaxy And Mass Assembly (GAMA) survey data for a highly spectroscopically complete set of low red- shift galaxies (Driver et al. 2011; Liske et al. 2015). We exploit, in particular, fits to the multi-wavelength spectral energy distributions (SEDs) made using stellar population synthesis techniques, which provide star formation rates (SFRs) and parameterised star forma- tion histories (SFHs). This information allows us to investigate, for example, the evolutionary stage reached by green valley galaxies, how their evolution has proceeded in the recent past, and how it is likely to proceed into the future. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | All magnitudes used in this work are in the AB system. Where relevant we use H0 = 70 kms−1Mpc−1, Ωm = 0.3 and ΩΛ = 0.7 as in Taylor et al. (2011), from whose (updated) catalogue we take the GAMA masses, rest frame magnitudes, dust-corrected colours and fitted stellar population parameters. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The GAMA survey provides a multi-wavelength database (Driver et al. 2011, 2016) based on a highly complete galaxy redshift sur- vey (Baldry et al. 2010; Hopkins et al. 2013; Liske et al. 2015) covering approximately 280 deg2 to a main survey magnitude limit of r < 19.8. This area is split into three equatorial (G09, G12 and G15) and two southern (G02 and G23) regions. The spectro- scopic survey was undertaken with the AAOmega fibre-fed spec- trograph (Saunders et al. 2004; Sharp et al. 2006) allied to the Two-degree Field (2dF) fibre positioner on the Anglo-Australian Telescope (Lewis et al. 2002). It obtained redshifts for ∼ 250, 000 targets covering 0 < z < 0.6 with a median redshift of z (cid:39) 0.2 with high (and uniform) spatial completeness on the sky in all the GAMA areas (Robotham et al. 2010; Liske et al. 2015). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Built around the redshift survey, photometric data are pro- vided at a wide range of wavelengths from the far-ultraviolet to far-infrared. Full details can be found in Driver et al. (2011, 2016), Liske et al. (2015) and Baldry et al. (2018). In this work we utilise the full GAMA II sample from the equatorial regions. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | extinction curve. A total of 195669 galaxies have high quality spec- tra (nQ > 2 in the GAMA catalogues) and fitted stellar masses (in the catalogue StellarMassesv19). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The current investigation uses a sample of GAMA galaxies tuned for our specific priorities. The primary sample is seelected from the three equatorial GAMA regions and limited to the redshift range 0.1 < z < 0.2. It is essentially volume limited for moderate to high luminosity galaxies (Mr < −20), and samples down to Mr = −18.5 at the low z end. This overall sample contains 62188 galaxies. As in Paper I, the limited (fairly low) redshift range means that there is only a small differential in look-back time (∼ 1 Gyr), and hence potential evolution, within the sample. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | To be conservative, we again follow Paper I and further restrict our samples to include only systems with r-band derived axial ra- tios b/a > 0.5, in order to exclude galaxies with edge-on disks. These potentially have less reliable intrinsic photometry, given the large and possibly more uncertain dust corrections which are made necessary by strong inclination dependent obscuration (e.g. Driver et al. 2007). The adopted ‘screen’ model of the dust extinction as used in Taylor et al. (2011), and most other similar work, will also be less appropriate for stars in edge-on disks (Disney et al. 1989; Baes & Dejonghe 2001). This leaves 39154 ‘round’ galaxies be- tween z = 0.1 and 0.2. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | In addition, as a check on any mass dependent effects, we also select a mass limited subsample of these galaxies which have stel- lar masses in the range 10.25 < log(M∗/M(cid:12)) < 10.75, the same range as used in Paper I. Finally, in Section 3.5, we briefly compare to a more local sample of 805 objects at 0.02 < z < 0.06, other- wise limited in similar ways to our main sample (see Kelvin et al. 2018, henceforth Paper II). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | As in Paper I, the main sample is then split into three colour- selected subsamples – ‘red’, ‘green’ and ‘blue’ – in terms of their intrinsic stellar, i.e. dust corrected, rest-frame, (u∗ − r∗) colours. Specifically, we divide our sample into these three broad colour bins based on the surface density of points in the intrinsic (u∗ − r∗) colour-mass plane, utilising the colours of the best-fitting SED (Taylor et al. 2011). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | We should note that, while our chosen colour is clearly sen- sitive to the true evolutionary state of the stellar population, as re- quired, the specification of a ‘green valley’ in general terms is not clear cut. The original definition came from GALEX observations in the ultraviolet, as discussed by Wyder et al. (2007), Martin et al. (2007) and Salim et al. (2007), amongst others, and selecting in terms of different optical or ultraviolet colours will likely produce different ‘green’ samples (see Salim 2014). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Nevertheless, intrinsic (u∗ − r∗) colours do allow us to iden- tify objects with stellar populations intermediate between purely passive and those undergoing so-called “main sequence” star for- mation (Noeske et al. 2007; Speagle et al. 2014) which dominate in the blue cloud. Evidently, any galaxy transitioning away from the main sequence of star formation must pass through this green colour regime at some point. In addition, contamination by dusty star forming galaxies, which could appear in samples defined only by observed colours (e.g. S´anchez Almeida et al. 2010; Casado et al. 2015) should be minimised. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 1. The stellar mass versus intrinsic stellar population (u∗ − r∗) colour plane for all non-edge-on (‘round’), 0.1 < z < 0.2 GAMA galax- ies. The division into ‘red’, ‘green’ and ’blue’ subsets (plotted in the ap- propriate colours) is as in Paper I, but extended to a wider mass range (see text). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | and we merely extend it to higher and lower masses. Blue galaxies are defined as those bluer than a line connecting (u∗ − r∗) = 1.3 to 1.4 over the same mass range, again covering the parameter space where sources typically classified as blue cloud galaxies reside. Galaxies are denoted as green between these two lines. The mid- point of these two lines essentially follows the minimum surface density of objects in colour as a function of mass, i.e. the centre of the ‘green valley’ as normally defined for these bands. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Our overall colour-mass plane is shown in Fig. 1. Our blue, green and red subsets contain 21406, 5068 and 12628 galaxies, re- spectively (ignoring a handful of objects outside the plotted limits). Histograms of the colour distributions for the red, green and blue galaxies are shown in Fig. 2. There is a small amount of overlap because of the mass dependence of the chosen edges of the green valley. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Note that the mass distribution, spanning 108 to 1012.5M(cid:12), varies between subsamples exactly as expected, with the blue frac- tion increasing towards lower masses and the red fraction towards higher masses (e.g. Kauffmann et al. 2003a,b; Baldry et al. 2004, et seq.). We have checked that this is not an artefact (circularity) of the method for determining M∗ based on the SED (and therefore colour), by alternatively plotting colour against K−band luminos- ity (a good proxy for mass; Bell et al. 2003), obtaining entirely equivalent results. The wide mass range accounts for the lower green galaxy fraction here (13%) than in Paper I ((cid:39) 20 %), which concentrated on a narrow intermediate mass range. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | While our colour cuts can be considered somewhat arbitrary – as noted above there is no single definition as to what consti- tutes a green valley galaxy – as in Paper I we will also consider galaxy properties across the continuum of intrinsic colours and use the three broad colour bins for primarily descriptive purposes. This is similar to the approach of Oemler et al. (2017) who divide their galaxies into subsamples via their sSFR values. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 2. Histograms of the (u∗ − r∗) colours for each colour subsample. The subsamples overlap slightly because of the slope of the green valley in the colour-mass plane. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | paper (again following Paper I), we utilise the SFRs obtained by use of the MAGPHYS code (da Cunha, Charlot, & Elbaz 2008) as adapted for use with GAMA (see Driver et al. 2016; Wright et al. 2016). Further, there are measures of the SFH of each galaxy. These are again obtained from the multi-band SED via a simple parame- terized fit (Taylor et al. 2011), as discussed below. An alternative, and complementary, route towards the stellar population parame- ters, and specifically the SFH, would be via fitting models to the galaxy spectra (e.g. Tojeiro et al. 2007; Wilkinson et al. 2017, and references therein), but this is beyond the scope of the present pa- per. In any case, as we are interested in galaxies subdivided by their broadband colours, it is more direct for our purposes to use models fitted to their SEDs. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | For orientation, we note that the majority of our green sample have a MAGPHYS-determined sSFR averaged over the last 108 yr 1) of 10−11 < sSFR8 < 10−10.2 yr−1 (which we label sSFR8 (so, e.g., for a green galaxy of observed stellar mass 1010.6M(cid:12) we have a SFR (cid:39) 1M(cid:12)/yr), with a tail to lower values. As ex- pected, most blue galaxies have significantly higher sSFRs (up to sSFR8 ∼ 10−9.4yr−1) and the red galaxies mostly have sSFR8 < 10−10.8yr−1 (see Fig. 3). In addition, we can see that the range of green valley sSFRs is largely independent of mass (Fig. 4). The overall (bimodal2) distribution of sSFR for GAMA galaxies, and its dependence on environment, was discussed in Davies et al. (2016a). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 3. Histogram of specific star formation rate for the three colour sub- samples. Here, the star formation is the value averaged over the last 108 years as derived via MAGPHYS (see Driver et al. 2016). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 4. Distribution of specific star formation rate (defined as in Fig. 3) as a function of stellar mass, for the green galaxies only. Note the peak at around sSFR8 (cid:39) 10−10.6yr−1 at all masses. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | epoch of observation (the ‘age’ of the galaxy3). The Taylor et al. modelling also assumes stellar population evolution models from Bruzual & Charlot (2003) with a Chabrier (2003) stellar initial mass function, a variety of possible metallicities, and the Calzetti extinction law, noted earlier, with E(B-V) as another variable in the fit. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | While other broad types of SFH are obviously possible, such as rising or constant SFR prior to the exponential decay (e.g. Lee et al. 2010; Schawinski et al. 2014; Smethurst et al. 2015; L´opez Fern´andez et al. 2018), Belli et al. (2018) have shown that sim- ple exponential SFR decays – ‘tau models’ – are generally entirely adequate to characterise the SFH of most galaxies and give equiva- lent results to more complex models in terms of derived parameters such as the median stellar population age (see also Chilingarian & Zolotukhin 2012). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | In addition, it is easy to see that such a SFH arises naturally in simple ‘closed box’ models of galaxies with a Kennicutt-Schmidt (Schmidt 1959; Kennicutt 1989) type star formation rate propor- tional to gas content or, indeed, similar but more general models with inflow and outflow proportional to the SFR (e.g. Tinsley & Larson 1978; see Trussler et al. 2018 for a recent discussion in the context of quenching). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Obviously, the ‘tau model’ does not take account of any sub- tleties in the SFH such as late time bursts of SF, though this is expected to be significant mainly in low mass galaxies below our sample limit. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The modelling also assumes that the whole galaxy can be rep- resented by a single SFH, when in fact in a galaxy containing both a bulge and a disk one might expect two distinct stellar populations formed on different timescales to contribute to the overall colours (e.g. Lagos et al. 2018; L´opez Fern´andez et al. 2018, and references therein). Finally, as is well known, any parameterisation based on observed optical colours will, for strongly star forming galaxies, be weighted towards recently formed stellar populations. Luminos- ity weighted ‘ages’, in particular, will be less than the actual time over which stars have formed. Nevertheless, the adopted model fits give consistent parameterisations of SFHs suitable for the present analysis. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | We note first (Fig. 5) that there is only a small difference in age between green and red galaxies. Green galaxies did not form the bulk of their stars much more recently than reds; in both subsam- ples most galaxies are modelled as starting – or, more realistically, peaking – their star formation ∼ 7 Gyr before the epoch of obser- vation, i.e. ∼ 9 Gyr ago (z ∼ 1.5), again essentially independently of mass (Fig. 6). Unsurprisingly, given that recent star formation tends to dominate the colours, most blue galaxies are modelled as having significantly smaller ages (even at the same mass), and the youngest galaxies are those of the lowest mass. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 5. Normalised histograms of fitted ages, t, for each colour subsam- ple. Note the strong peak at around 7 Gyr for both red and green galaxies. There is very little overlap between the red and blue samples. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 6. Distribution of age, t, as a function of stellar mass for each colour subsample. Note the independence of age on mass for the red and green samples. Blue galaxies, on the other hand, are typically younger at lower masses (‘downsizing’). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 7. Normalised histogram of fitted e-folding times, τ , for each colour subsample. Green galaxies are clearly at intermediate values compared to the other subsamples. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | SFR e-folding times typically ∼ 2 Gyr. These values are evidently somewhat shorter than ours from Taylor et al. (2011). However, the two estimates cannot be directly compared, as Smethurst et al. assume a constant SFR for several Gyr prior to the exponential de- cline; clearly a long constant period followed by a rapid decline would be modelled as a slower overall decline in our method, so we would in general expect our τ to be longer than theirs, as is the case. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | On the other hand, from a detailed study of a relatively small number of galaxies, with a range of colours and types, from the CALIFA survey, L´opez Fern´andez et al. (2018) find that their summed stellar populations (so effectively a galaxy of intermedi- ate colour) can be modelled via a SF e-folding time of 3.9 Gyr, en- tirely consistent with our values here. They also find shorter times for early type galaxies (effectively our reds) and longer times for late type galaxies (our blues). They further note that the overall fall-off is very similar to the rate of decline of the cosmic star for- mation density (the cosmic star formation history) as measured by, e.g., Madau & Dickinson (2014) and Driver et al. (2018). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Finally for this section, we consider the ratio of t to τ , i.e. the number of star formation e-folding times which have passed since the peak in star formation. Fig. 9 shows clearly that green galaxies are intermediate between red and blue in this measure of evolution, and moreover that all green galaxies have very similar values of this parameter, with a tight concentration of values around t/τ = 1.6). The distribution is significantly narrower than that in sSFR itself from Fig. 3. Nevertheless Fig. 10 shows that there is some remain- ing small mass dependence in t/τ , with a tendency towards slightly higher values (further advanced evolution) for the more massive green galaxies. This presumably arises from the somewhat redder colours selected for the green valley at higher mass. It could possi- bly also reflect an environment dependence, as more massive galax- ies are typically in larger groups. We defer any further discussion of the environment of green galaxies to a subsequent paper. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 8. Distribution of fitted star formation e-folding times, τ , as a func- tion of stellar mass for each colour subsample. Note the significant depen- dence of τ on mass for the red and green samples, while blue galaxies have similar (large) τ at all masses. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 9. Normalised histogram of fitted ages in units of e-folding time, t/τ , for each of the colour subsamples. The green galaxies show a distinctly different distribution from the others, with values concentrated around t/τ (cid:39) 1.6. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | those galaxies fitted by an old age t but an intermediate timescale τ and with a narrow distribution of t/τ . |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The above provides, at least qualitatively, the answer to our first question, i.e. it does appear that green galaxies are at a specific characteristic point in their evolution, where the star formation has decreased by a factor (cid:39) 5 relative to its peak value. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 10. Distribution of fitted ages in units of e-folding times, t/τ , as a function of stellar mass for each colour subsample. While green galaxies all have closely similar values, there is nevertheless a slight increase with mass, i.e. higher mass green galaxies are further through their evolution. This is also true (more strongly) for the red and blue samples. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Notice that this implies that as galaxies of any given τ age, they move along the same overall relation. This is evident from the bottom panel of Fig. 11, since the different colour subsamples have different typical values of τ (and is easily checked from subsamples of fixed τ ). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The close correlation then implies that we can empirically es- timate how the colour should evolve over time, viz. a change by (cid:39) 0.4 magnitudes as t/τ changes by a factor 2. We should note here that this correlation only applies in the plotted range where colour is an essentially linear function of evolutionary state. Out- side this range the colours vary less with evolutionary state (bend- ing to a slope of (cid:39) 0.8 rather than the 1.3 in the above equation; see the top panel in Fig. 11) as they asymptote to those of totally star forming or totally dead stellar populations (cf. Oemler et al. 2017; Eales et al. 2017). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | for each of our colour subsamples. We can note the strong resem- blance to the original (actual) colour distribution in Fig. 2. We therefore use the empirically predicted evolution of this pseudo- colour as a proxy for the evolution of the real colour. We return to this below. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 11. Variation of (u∗ − r∗) colour against fitted age in units of e- folding times, t/τ (top), and relative to the overall correlation (bottom), for the mass limited subsample. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 12. Histograms of the pseudo-colour (u∗ − r∗)(cid:48) (see text) for each colour subsample of our main sample. Note the strong resemblance to Fig. 2, showing the actual (u∗ − r∗) colours. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | the absence of any significant change in stellar mass) by a factor ∼ 8. Further, from Fig. 7, green galaxies have τ ∼ 4 Gyr. Match- ing this to the assumed SFR decline e−∆t/τ , then ∆t/τ (cid:39) ln 8, i.e. ∆t ∼ 8 Gyr. Thus despite various other similarities, noted here (age), in Paper I (S´ersic index, bulge-to-disk mass ratio) and else- where (e.g. Fang et al. 2013, for surface density), it is evident that in global star formation terms green valley galaxies (with fitted ages ∼ 7 Gyr) are only ‘half way’ to becoming a thoroughly ‘red and dead’ early type (unless their SFR decline speeds up, of course). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | However, this is not the same as saying that galaxies stay green for this length of time. Leaving the green valley merely requires them to cross the border into red territory. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | If we take, as a specific example, galaxies ‘in the middle’ of the green valley to be within 0.02 magnitudes of the central colour (for their mass), then we can check that, similar to what was as- sumed above, they have a mean log(sSFR) (cid:39) −10.7 (and a mean τ of 4.0 Gyr, as for the whole green sample). If we similarly take the subset of galaxies redder than the edge of our green valley by less than 0.02 magnitudes4, we find that they have an average log(sSFR) (cid:39) −11.1, i.e. a factor 2.5 lower. Thus a typical green will take ∼ 3 − 4 Gyr to (just) become a red galaxy or, rather, to reach the same sSFR as a current red galaxy (we consider actual colour evo- lution in the next section). We can think of this as an estimate of the ‘half-life’ of a current green galaxy; after this time the redder/lower sSFR half of them (at a given mass) will have evolved sufficiently to move out of the typical range of sSFRs observed for current green valley galaxies. Note that for simplicity, throughout the paper we will refer to our 0.1 < z < 0.2 samples as the “current” green valley, etc., to distinguish this from higher redshift studies, though strictly the galaxies are observed as they were 1-2 Gyr ago. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | We can extend this to estimate how long a galaxy at the blue side of the green valley will take to cross to the red side. Galaxies within 0.02 magnitudes of the blue edge of the green valley have a mean log(sSFR) (cid:39) −10.5 and mean τ (cid:39) 4.7 Gyr. For one to reach the middle of the green valley, the sSFR must change by a factor 1.6, which will take ∆t ∼ 2 Gyr. Hence, to complete the crossing of the green valley from blue to red should take a total of ∼ 5 − 6 Gyr. Notice that the middle of the valley is rather nearer to the blue edge than the red edge in star formation terms (factor 1.6 rather than 2.5) and this more than makes up for the slightly longer e-folding times of the bluer galaxies, making the first half of the transition faster overall. We should not necessarily believe the details of these estimates, of course, as we should really allow for the spreads in sSFR (or colour) and τ . In particular, galaxies (of any colour) with smaller τ will change their sSFR and colour more quickly. In addition, as seen earlier, τ and sSFR both depend on mass and there is a sizeable scatter in sSFR at a given colour. We therefore make more detailed calculations in terms of the evolution in colour in the following section. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | It is worth noting that for the simple case of a closed-box model with SFR proportional to gas mass – which results in an exponential fall-off, as assumed in our analysis – τ is equal to the gas depletion timescale derived simply as the remaining gas mass divided by the current SFR (Roberts 1963). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Notice that point (ii) above reflects that there may be a cer- tain inconsistency in our simple approach, in that we assume that τ remains fixed for each half of the transition (blue to green and green to red), while we know that blues have longer τ than greens which in turn have longer τ than reds, perhaps suggesting a steady shortening of the timescale while traversing the valley (though see Section 3.5 for arguments against this). Either way, the above argu- ments should give us a plausible first estimate without requiring us to model a more complex form for the SFH. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | With respect to numbers, if we do split the evolution into two phases, then from our calculation we can say that a galaxy in the redder (lower sSFR) half of the green valley galaxy should have es- caped from the valley (in SF terms) in 3−4 Gyr time. Over the same timescale, some current blue (higher sSFR) galaxies should enter. Looking at this purely from the sSFR (we return to the colours in the next section), the blue galaxies which can enter the green valley in the next 3 Gyr are evidently those which are sufficiently close to the green valley boundary that with their combination of sSFR and τ , schematically, sSFRbluee−3/τblue → sSFRgreen. If the number in this ‘pre-transition’ region is a suitable fraction of all blue galax- ies then we can repopulate the green valley at an appropriate rate to keep a sensible fraction of galaxies in the green valley at all times (see e.g. Paper 1 and, at higher redshifts, Cooper et al. 2006; For- rest et al. 2018). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Working in sSFR itself is problematical, though, as green galaxies are spread over a significant range (∼ 1 dex) of sSFR and at almost all sSFR values, even at the centre of the valley, there are actually more red and blue galaxies at a given sSFR than green (see Fig. 3). We show some plots of evolved sSFRs themselves in the Appendix, but in this section and the rest of the main body of the paper we concentrate on the colours, or the pseudo-colours (u∗ − r∗)(cid:48) from Section 3.2. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | We can easily evolve the distribution of pseudo-colours by adding any chosen ∆t to the fitted ages of our galaxies and re- calculating (u∗ − r∗)(cid:48). As examples, the middle and bottom panels of Fig. 13 show the evolved pseudo-colours for ∆t = 2 Gyr and 4 Gyr, compared to the observed distribution (cf. Fig. 12) in the top panel. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Therefore to get a more accurate picture, in the middle panel of Fig. 14 we now show the whole evolved pseudo-colour versus mass plane after 2 Gyr. (The top panel shows the observed distri- bution). Note that we use the mass at the epoch of observation in all the plots, and do not attempt to account for any, generally small, changes of stellar mass with time. Of the ∼ 5000 green galaxies, ∼ 2800 have crossed the (current) green valley upper border, to become red, if not yet dead. On the other hand, of course, a number of formerly blue galaxies have ‘moved in’. Indeed, because of the large initial population of blues, this more than compensates and we have a total of ∼ 9500 galaxies in the area of the current green valley. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | After 4 Gyr ( Fig. 14 bottom panel) all except a handful of the high mass galaxies among the current green galaxies sample have moved out of the green valley as currently defined (its top edge is indicated by the diagonal line). Indeed this is actually largely true by 3 Gyr. These (formerly) green galaxies have been replaced (in the current valley) by a large number (∼ 15000) of the current blue galaxies. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | These timescales also apply for the mass-limited sample from Paper I. The most important difference for this subsample com- pared to our main one is the lack of very blue galaxies and partic- ularly the corresponding decrease in the fraction of galaxies with very short t/τ (see Fig. 10). However this has essentially no effect on the evolution across the green valley. We still remove of order half the current green galaxies in ∼ 2 Gyr and essentially all of them in ∼ 4 Gyr (Fig. 15), just as for the complete sample. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Of course, all these estimates are somewhat dependent on the choice of green valley galaxies. If we had chosen narrower colour limits, then clearly galaxies would be able to transit the green colour range in a shorter time. In addition, we can see that the tran- sit times are mass dependent. From the middle and bottom panels of Fig. 14, we can see that the lowest mass green galaxies have just crossed the (diagonal) top edge of the current valley after 2 Gyr, while the highest mass objects only reach it after 4 Gyr. This is de- termined not only by the mass dependent SFH timescales (Figs. 9 and 10), but by the wider range of ‘green’ colours at higher mass in our original selection (since the upper and lower colour limits are not parallel). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Recall that we have used a fixed slope for the evolution of the colour with log(t/τ ) for all our galaxies, so strictly we have over- estimated the shift of the colours for the galaxies with the longest and shortest t/τ values (bluest and reddest galaxies) since these evolve more slowly in colour, as noted earlier. Thus the top of the red distribution and the bottom of the blue distribution should re- ally be moved downwards slightly in the two lower panels of Fig. 14, but this has no effect on the green galaxies (present or near future), which are well within the linear regime of our empirical colour correction in Fig. 11. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 13. Histograms of pseudo-colour (u∗ − r∗)(cid:48) for the colour subsets: (top) as observed (as Fig. 12), (middle) as evolved forwards by 2 Gyr and (bottom) as evolved forwards by 4 Gyr, assuming a constant τ for each galaxy (see text). The total distributions are given by the grey histograms. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 14. Distribution of pseudo-colour (u∗ − r∗)(cid:48) as a function of (ob- served) mass, (top) as observed, (middle) as evolved forwards by 2 Gyr and (bottom) as evolved forwards by 4 Gyr (see text). The diagonal black lines in the middle and bottom panels represent the red edge of the (true) colour green valley from Fig. 1, demonstrating that virtually all current green galaxies have evolved out of the current green valley after 4 Gyr. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 15. Distributions of pseudocolour as observed (top) and evolved forward for ∆t = 2 and ∆t = 4 (middle and bottom) for the mass limited sample. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | half to all of the current green galaxies out of the current valley will take 1.5 to 4 Gyr. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The timescales derived in this section differ by a factor ∼ 1.5 from the earlier estimate of the green valley crossing time based on ‘typical’ sSFRs of green and red galaxies, primarily because of the wide spread of sSFR at a given colour and the more precise consideration of the green valley boundaries. Since it is the change in colour that we are specifically interested in here, re. the definition of ‘green valley’, the lower value of the timescale found from the galaxy by galaxy calculation should provide the better estimate. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | As a sanity check, we can explore briefly the changes in colour across the redshift range surveyed by GAMA. To do this we choose an additional sample from the redshift range 0.02 < z < 0.06 (cf. Moffett et al. 2016), but with all other constraints the same as in our main sample. In order that mass is not a complicating factor, we choose only galaxies with 1010 < M∗/M(cid:12) < 1011. (Using the full set of data available at low z would include disproportion- ally large numbers of low luminosity blue galaxies compared to our main sample). Notice that this is a wider mass range than for our previous mass limited sample (and that in the low z green val- ley study of Paper II), but the narrower range would leave too few low z galaxies for a statistical study. With the wider limits we have 805 objects. To obtain the longest possible time difference, we then choose a sample in the same mass range and at 0.19 < z < 0.20 from our main sample. This contains 2996 objects. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The look-back time difference between z = 0.02 and z = 0.2 is just over 2 Gyr, so would expect to see comparable differences in the samples to those illustrated in, e.g., Fig. 13. In this case we can use the actual (as opposed to pseudo-) colours, and we find that indeed, the overall distribution is shifted redwards by around 0.15 magnitudes in the low z sample, as shown in Fig. 16, consistent with the predictions of our modelling (compare the top and middle panels of Fig. 13). The values of t/τ for the galaxies in the new valley are also increased by ∼ 0.5, as expected if τ has remained typically 4 Gyr for these objects. Inter alia this supports the physi- cal reality of the derived parameters which we use. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Indeed, the close match of the distribution of the predicted evolved colours of the 0.19 < z < 0.20 sample to the actually ob- served colour distribution at 0.02 < z < 0.06 is strong empirical evidence for evolution with fixed τ in each galaxy, i.e. subsequent evolution of the z = 0.2 galaxies following the same fall-off as in their preceding SFH. If (z = 0.2) green galaxies suddenly acceler- ated their evolution in the last 2 Gyr we would expect to see more red galaxies, and even fewer in the valley, than is actually the case at low z. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Fig. 17 illustrates this further. Given that the low z sample do not all correspond to a single epoch, we have here evolved the ob- served colours of the galaxies in both the 0.02 < z < 0.06 and the 0.19 < z < 0.20 samples to redshift zero via the look-back times to their individual redshifts. The agreement is very good (except for the bluest galaxies, for which we know we have overestimated the colour change, effectively moving the blue peak one bin too far to the right in the plot), giving further evidence that evolution of the the colours with a fixed τ for each galaxy gives a consistent re- sult. A greater amount of evolution for the z (cid:39) 0.2 green galaxies, because of an accelerated decline in SF over the last 2 Gyr, would generate colours which would tend to be too red and, as noted pre- viously, produce too large a dip at green colours. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 16. Normalised distributions of observed (u∗ − r∗) colour for sub- sets of galaxies at 0.19 < z < 0.20 (light blue) and 0.02 < z < 0.06 (grey), as described in Section 3.5. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 17. Normalised distributions of (u∗ − r∗) colour for the subsets of galaxies at 0.19 < z < 0.20 (light blue) and 0.02 < z < 0.06 (grey) evolved to z = 0 via their individual look-back times. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The predicted distribution of what we should see at later times also suggests that we ought to consider whether to allow for the fact that ‘a’ green valley, as defined as a dip (or plateau) in the colour distribution, may move redwards as the overall sSFR distribution declines. Does the same range of colour or t/τ represent a valley at all epochs? Eventually, presumably, galaxies may reach the current green valley colour while still being amongst the bluest objects. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Looking for a dip in the pseudo-colour distributions, or at least the edge of the blue peak, we can see from Fig. 13 that at ∆t = 2 Gyr this will be at roughly (u∗ − r∗)(cid:48) (cid:39) 1.75 (compared to (cid:39) 1.55 currently), and by ∆t = 4 Gyr it is at (u∗ −r∗) (cid:39) 1.85. As noted above, very red and very blue galaxies do not follow the same t/τ versus colour trend that we have utilised in our evolution, since the colours pile up at the ends of the colour distribution regardless of the detailed SFH (cf. Eales et al. 2017, 2018; Oemler et al. 2017). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Allowing for this does not change our conclusions regarding the evolution of green galaxies in the next few Gyr, though a pile up at the reddest colours would emphasise the dip at green colours. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | By ∆t = 4 Gyr, t/τ for typical current green galaxies will be 1.6 times its current value (as usual, assuming no change in τ ), also corresponding to (u∗ − r∗)(cid:48) (cid:39) 1.7. In that sense the cur- rent green galaxies are somewhat ‘marooned’ in an apparent in- termediate state as all three current colour subsets will evolve to redder colours. Unless they change their τ , current green galaxies will always have sSFR in between those of the current red galaxies with little star formation and short τ and current blue galaxies with significant star formation and long τ , even though they will have moved out of the current green valley. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | As they do evolve more than the reds in colour, they will even- tually close up on the existing red galaxies, once their SFR be- comes negligible, and essentially merge into the red sequence, all of whose galaxies will be asymptoting to late time old stellar popu- lation colours (Eales et al. 2017). How long this takes will evidently depend on how closely similar in colour we require galaxies on the red sequence to be. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Reversing this and looking back in time, the same arguments would suggest that the valley between strongly and weakly star forming galaxies should have been at bluer colours. This is in agreement with, for instance, the data from Bell et al. (2004) who show that in COMBO-17 the colours of the blue peak, adjacent to the valley, and of red sequence galaxies both move redwards from z (cid:39) 1 towards the present day (see also Belli et al. 2018). We will explore running the SFH models backwards and comparing to higher redshift data samples in a separate publication. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | The primary outcomes of the above analysis are that current green valley galaxies largely share a common SFH and have reached a specific phase of their evolution in terms of the fall-off in their SFR. In terms of timescales, we find that (current) green galaxies can leave/cross the (current) green valley on timescales of 1.5- 4 Gyr. This is similar to, but somewhat longer than the 1-2 Gyr timescale proposed, on the basis of the fraction of the population residing in the green valley, in Paper I, and comparable or slightly longer than the 2.6 Gyr suggested by Rowlands et al. (2018) from an analysis of spectroscopic indices of candidate transition objects. However, interestingly as we have no specific quenching events, it is very similar to or shorter than typical estimates of the timescale for the fading which follows the cutting-off of gas sup- plies (‘strangulation’ or ‘starvation’) of star-forming galaxies, gen- erally between 4 and 8 Gyr (Bekkii et al. 2002; Peng et al. 2015; Wheeler et al. 2014; Guo et al. 2017; Trussler et al. 2018). The larger of these estimates are, rather than the green valley crossing time, more comparable numerically to the e-folding times of the SFR that are measured here for green (blue) galaxies of τ ∼ 4 (6) Gyr, or indeed the ages (by our definition) of green galaxies, t ∼ 7 Gyr. Indeed, it should be noted, that different authors mean different things by ‘quenching time’, for instance the time to reach the green valley or reach the red sequence, the time since SF started to decline from some earlier constant level, or an e-folding time of the SF (within various SFH models). Clearly, each may result in quantitatively different times being deduced. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | colour subsamples for instance, remain true even if we substantially change our priors, and a strong correlation between colour and t/τ is robustly obtained. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | As noted earlier, in a simple closed box (presumably appro- priate after infall is prevented, for instance) with an exponentially declining SFR, τ is equal to the time for the remaining fuel to be used up at the current rate (e.g. Hopkins et al. 2008). Clearly direct measurements of the gas masses for our galaxies would be very useful to test the overall picture (cf. Catinella et al. 2012, for the GASS survey). Upcoming surveys such as DINGO (Duffy et al. 2012) will provide this data. Given that the typical t/τ implies that current green valley galaxies have reduced their SFR by a factor around 5 since its peak, then in a simple Kennicutt-Schmidt type model, we would also expect the gas supply to have decreased by a similar factor. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | On a similar timescale to galaxies leaving the valley, we might expect current blue galaxies to repopulate it. However, it is appar- ent that most blue galaxies have longer τ than green valley galaxies so would not replicate the (current) green galaxy evolutionary pa- rameters even after fading, unless only those blue galaxies with τ at the short end of their range evolve to be green. However, due to the increased t they can have similar t/τ to current green galaxies. We should stress that the timescales calculated throughout this paper depend on the assumed exponentially declining model SFH with a fixed τ for each galaxy. However, we can note that even changing the typical τ of green galaxies to that observed in red galaxies does not grossly change the results (giving values at the low end of the range we quote), and that furthermore there is no evidence in the comparison with the low z colour distribution for any systematic speeding up of the evolution of the green galaxies over time. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | We should also remember that the present work largely refers to galaxies in the field and in low mass groups. There are no re- ally large clusters in the GAMA areas. The timescales are there- fore strictly only relevant to such environments. Within these, split- ting the galaxies into grouped and isolated (Robotham et al. 2011) we see virtually no differences in the τ and t/τ distributions for the green galaxies. Of course, it remains possible that galaxies in high density environments (which we do not sample here) evolve through their green phase more rapidly. We will explore this ques- tion elsewhere. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | In Paper I it was shown that the majority of green galaxies (in the mass limited sample) had underlying structural parameters consistent with those of red galaxies. In particular, fitting the ra- dial light profiles at red or near infrared wavelengths (effectively the mass surface density) by a single S´ersic (1968) component, the distributions of the index n were strongly overlapping for the two subsamples. Blue galaxies on the other hand had significantly dif- ferent structural parameters. It was therefore argued that most green galaxies could evolve directly onto the red sequence with no (fur- ther) change to their mass distributions. In particular, this was taken to support the view (e.g. Cheung et al. 2012; Fang et al. 2013; Tay- lor et al. 2015) that a significant spheroidal component needs to be in place before a galaxy becomes green. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 18. Distributions of τ and t/τ against Sersic index for blue and green galaxies in the mass limited sample. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | with strongly ongoing SF. However, the blue galaxies with large n, say above 2.5 (log n > 0.4), have on average shorter τ than the blue pure disks, but this is not by a large factor and is not enough to make them significantly overlap with the green population. Nev- ertheless, looking at the overall distribution for the combined blue and green populations does show a trend for shorter τ (and larger t/τ ) as nr increases, which is compatible with the effects of an in- creased bulge component changing both n and the overall SFH of a two component system (cf. L´opez Fern´andez et al. 2018). |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | This is supported in a general way by the results from the smaller low z sample, for which there exist morphological type classifications (Moffett et al. 2016, Paper II). Intermediate values of τ , around 4 Gyr, typical of our main green sample, are seen primarily in the S0-Sa and E classes, i.e. in those galaxies with a large bulge componen, rather than in the Sab-Scd class. This agrees with the CALIFA results of L´opez Fern´andez et al. (2018) who find that S0/Sa galaxies have intermediate SFH timescales, and of Bit- sakis et al. (2018) who show that CALIFA galaxies between the star forming main sequence and the quiescent region have morpho- logical classifications from elliptical to early spiral. Chilingarian et al. (2017) likewise find that S0 and Sa galaxies have typical colours which place them in their green valley region. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | galaxies may never leave their valley. Since red galaxies pile up at their asymptotic colours at large ages and (at least currently) the blue galaxies decrease their star formation more slowly than other types, the present green galaxies may always remain as an interme- diate population as the whole galaxy population reddens over time, at least until the point where they become completely red and dead, lose the imprint of their individual SFHs and merge into the red sequence. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | GAMA is a joint European-Australasian project based around a spectroscopic campaign using the Anglo-Australian Telescope. The GAMA input catalogue is based on data taken from the Sloan Dig- ital Sky Survey and the UKIRT Infrared Deep Sky Survey. Com- plementary imaging of the GAMA regions is being obtained by a number of independent survey programs including GALEX MIS, VST KiDS, VISTA VIKING, WISE, Herschel-ATLAS, GMRT and ASKAP providing UV to radio coverage. GAMA is funded by the STFC (UK), the ARC (Australia), the AAO, and the participating institutions. The GAMA website is http://www.gama-survey.org/. This work made extensive use of TOPCAT (Taylor 2005) software packages, which are supported by an STFC grant to the University of Bristol. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | For illustration, Fig. 19 shows the observed and evolved sSFR dis- tributions (Section 3.4) against observed stellar mass, i.e. not allow- ing for any increase in the stellar mass with time. This latter will only be significant for the bluest galaxies with nearly constant star formation rates as a function of time (long τ ) which could change their stellar masses by a few tenths of a dex over a 4 Gyr time span. As expected, over time the current green galaxies decrease their typical sSFRs to ∼ 10−11yr−1, with an extension down to ∼ 10−12, the majority continuing to lie at the edge of the blue cloud in SFR terms. |
Processed_Galaxy_And_Mass_Assembly_(GAMA):_Timescales_for_ga.txt | Figure 19. Distribution of sSFR as a function of (observed) mass, (top) as observed (cf. Fig. 4), (middle) as evolved forwards by 2 Gyr and (bottom) as evolved forwards by 4 Gyr assuming an exponential fall-off with constant τ for each galaxy (see text). |
Processed_Gas_and_Star_Formation_in_the_Circinus_Galaxy.txt | ABSTRACT We present a detailed study of the Circinus Galaxy, investigating its star formation, dust and gas properties both in the inner and outer disk. To achieve this, we obtained high-resolution Spitzer mid-infrared images with the IRAC (3.6, 5.8, 4.5, 8.0 µm) and MIPS (24 and 70 µm) instruments and sensitive H i data from the Australia Telescope Compact Array (ATCA) and the 64-m Parkes telescope. These were supplemented by CO maps from the Swedish-ESO Submillimetre Telescope (SEST). Because Circinus is hidden behind the Galactic Plane, we demonstrate the careful removal of foreground stars as well as large- and small-scale Galactic emission from the Spitzer images. We derive a visual extinction of AV = 2.1 mag from the Spectral Energy Distribution of the Circinus Galaxy and total stellar and gas masses of 9.5 × 1010 M⊙ and 9 × 109 M⊙, respectively. Using various wavelength calibrations, we find obscured global star formation rates between 3 and 8 M⊙ yr−1. Star forming regions in the inner spiral arms of Circinus, which are rich in H i, are beautifully unveiled in the Spitzer 8 µm image. The latter is dominated by polycyclic aromatic hydrocarbon (PAH) emission from heated interstellar dust. We find a good correlation between the 8 µm emission in the arms and regions of dense H i gas. The (PAH 8 µm) / 24 µm surface brightness ratio shows significant variations across the disk of Circinus. |
Processed_Gas_and_Star_Formation_in_the_Circinus_Galaxy.txt | The Circinus Galaxy (HIPASS J1413–65), or Circinus in short, is a nearby, highly obscured spiral galaxy which was discovered by Freeman et al. (1977). It is located behind the Galactic Plane (b ∼ −4◦), hidden behind a veil of dust and high star density. Freeman et al. (1977) estimated a Holm- berg radius of ∼17′, a morphological type of Sb to Sd and a distance of around 4 Mpc. |
Processed_Gas_and_Star_Formation_in_the_Circinus_Galaxy.txt | Circinus hosts an active galactic nucleus (AGN), super- luminous H2O masers, a recent supernova, a star-forming nuclear ring, a spectacular pair of radio lobes and a gigantic hydrogen disk. In the following we briefly describe these and other features before focusing on the inner disk of Circinus where Spitzer data reveal the stellar disk and the onset of two symmetric spiral arms as never seen before. |
Processed_Gas_and_Star_Formation_in_the_Circinus_Galaxy.txt | from the nuclear region suggests high starburst activ- ity (Ghosh et al. 1992). Subsequent spectral line observa- tions by Oliva et al. (1994, 1998) confirmed a Seyfert 2 nucleus. The detection of a circum-nuclear ring in the Hα line is reported by Marconi et al. (1994) (see also Koribalski 1996; Elmouttie et al. 1998). Circinus also hosts highly variable H2O megamasers (Gardner & Whiteoak 1982; Whiteoak & Gardner 1986) which appear to arise in the circumnuclear accretion disk and in a low-velocity outflow (McCallum et al. 2009). Bauer et al. (2001) discov- ered a recent supernova, designated SN1996cr and located just 30′′ south of the nucleus of Circinus, and present a multi-wavelength analysis of its lightcurve. High-resolution Australia Telescope Compact Array (ATCA) radio contin- uum images reveal a prominent pair of linearly polarized radio lobes orthogonal to the disk (Elmouttie et al. 1998; Wilson et al. 2011). |
Processed_Gas_and_Star_Formation_in_the_Circinus_Galaxy.txt | Low-resolution 21-cm observations with the 64-m Parkes radio telescope reveal the enormous H i envelope of the Circinus Galaxy. Its large-scale gas disk extends over an area more than one degree in diameter and is much larger than the stellar disk (Mebold et al. 1976; Freeman et al. |
Processed_Gas_and_Star_Formation_in_the_Circinus_Galaxy.txt | The H i physical diameter of the Circinus Galaxy is sim- ilar in size to that of the M 83 galaxy (Koribalski 2008). While Circinus is relatively isolated and mostly undis- turbed, M 83 is surrounded by numerous dwarf galaxies, e.g. NGC 5253 (L´opez-S´anchez et al. 2012), and clearly affected by its environment. The tidally distorted outer H i arm and associated debris (close to NGC 5264) as well as a faint stellar stream provide evidence for gas accretion and weak gravitational interactions (Koribalski et al. 2012, in prep.). The large H i disks of Circinus and M 83, which are typical for their H i mass (Broeils & van Woerden 1994), allow us to measure their dark matter content out to radii of ∼50 kpc. High-resolution GALEX images of M 83 reveal an extended UV- or XUV-disk, suggesting that star formation is not re- stricted to the inner stellar disk (Thilker et al. 2005). High- density H i emission is an excellent tracer of star formation in the outer disk of galaxies; good examples are NGC 1512 (Koribalski & L´opez-S´anchez 2009), M 83 (Koribalski et al. 2012, in prep.) and Circinus (this paper). |
Processed_Gas_and_Star_Formation_in_the_Circinus_Galaxy.txt | Despite the richness of interesting features in the Circi- nus Galaxy, most studies have been limited to its nuclear region. While the 2MASS Large Galaxy Atlas (Jarrett et al. 2003) has provided an extensive view of the stellar con- tent of Circinus, the formidable dust opacity at 2 µm and the dense stellar foreground prevent a detailed investiga- tion of its large scale and low surface brightness disk. The mid-infrared (MIR) window is ideal to unveil both the old stellar disk population of the obscured Circinus Galaxy as well as the interstellar emission from the new born stars in the outer disk. The high angular resolution (2′′ in the IRAC 8 µm band) and sensitivity of the Spitzer telescope makes it more capable than previous infrared observatories (e.g., IRAS, ISO & AKARI). The on-board Spitzer high dy- namic range IRAC and MIPS imaging instruments allow us to study the full stellar extent of the Circinus disk and the distribution of its warm and cold dust. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.