Schawinski et al. Article - Evolution of Early and Late Types Through the Green Valley

Title: The Green Valley is a Red Herring: Galaxy Zoo reveals two evolutionary pathways towards quenching of star formation in early- and late-type galaxies

Authors: K. Schawinski, M. Urry, B. Simmons, L. Fortson, S. Kaviraj, W. Keel, C. Lintott, K. Masters, R. Nichol, M. Sarzi, S. Ramin, E. Treister, K. Willett, I. Wong, and S. Yi

First Author's Institution: Institute for Astronomy, Department of Physics, ETH Zurich

To read the full article, please click here.  

Article Summary:

The green valley has long been thought of as the crossroads of galaxy evolution – creating a divide between the star-forming galaxies of the blue cloud and passively-evolving galaxies of the red sequence. The intermediate color of green valley galaxies is thought of as an indicator that star formation in these samples was recently quenched, and studying this region between the two main populations will lead to a better understanding of the evolutionary pathways of galaxies, possibly even predicting what may happen to our own Milky Way. Galaxies spend most of their life on what is dubbed the Main Sequence, where there is a tight correlation between stellar mass and star formation rate (SFR).  It is believed that certain processes can cause them to leave the main sequence and begin to travel through the green valley. Most galaxies were presumed to follow a similar tract through this desolate region of color space, progressing through it in a relatively short timescale to keep the scarcity of galaxies observed in this region. In recent years, the Galaxy Zoo project has brought a new tool to the galaxy evolution table – a plethora of morphological classifications of galaxies imaged by the Sloan Digital Sky Survey (SDSS), allowing astronomers to inquire as to the effects of morphology on a galaxy's transition through the green valley. A recent study done by Schawinski et al. using data from the SDSS, Galaxy Evolution Explorer (GALEX), and Galaxy Zoo has found that different morphological characteristics will alter the movement of a galaxy through the green valley into its quiescent fate.

This study acquired a sample of mass-limited galaxies in the local universe from the SDSS, with redshifts ranging from z=0.02 to z=0.05.  To gain a better understanding of star formation histories, ultraviolet photometry from the GALEX was found for 71% of their sample. Galaxy Zoo classifications were used to determine the morphology of the sample, where morphology was assigned when volunteers agreed on the classification at a rate of 80% or more. This resulted in the classification of 18% early types, 34% late types, 45% intermediate types (galaxies that did not receive at least 80% votes for early or late type), and 3% merging, where these classification follow the morphology of Hubble's Tuning Fork (see figure 1). The science team believes the relatively high number of intermediate types most likely results from the abundance of systems in which the bulge or disk do not clearly dominate rather than poor imaging data.  

Figure 1. Example gri images ordered by Galaxy Zoo classifications.  The left, center, and right columns show blue cloud, green valley, and red sequence galaxies, respectively.  The top, center, and bottom rows show early-type, intermediate-type, and late-type galaxies, respectively.  

Though most late-type galaxies of the sample inhabited the blue cloud and most early types were in the red sequence, this study found that both early-type and late-type galaxies spanned almost the entire color range, and within a given morphological class, the green valley was nothing more than a collection of outliers. 

Figure 2. Reddening-corrected u-r color-mass diagram for the sample.
This figure shows two important findings - that both early and late type
galaxies span almost the entire color range, and that the green valley
is only defined in the all-galaxy color panel.  

The bimodality of galaxy colors is a result of the superposition of the two populations; late types are mostly in the blue cloud and decrease smoothly to the red sequence, and a few early types reach all the way to the blue cloud. Figure 2 shows a stellar mass-color contoured plot after dust extinction corrections, and indicates the position of the green valley in relation to the blue cloud and red sequence.   

Using ultraviolet and optical photometry, star formation rates (SFRs) were analyzed to find that early-type galaxies are quenched much more rapidly than their late-type counterparts. Late-type galaxies were still blue in the ultraviolet through the green valley, indicating that they were still undergoing star formation as they were being quenched. Transitions through the green valley were found to be highly dependent on morphology, with early-type galaxies exhibiting quenching on timescales as short as 250 Myr. If this process took longer, like the 1-3 Gyr track estimated for late-type galaxies, there would be a build-up of early-types in the green valley that are not accounted for in observations. Perhaps early-type galaxies transition through the green valley as fast as star formation would allow.

Figure 3. UV-optical dust corrected color-color diagrams of green valley
galaxies.  Color-coded evolutionary tracks using different quenching
 timescales are overplotted on the green valley early- and late-type plots.
Emission line-selected AGN in host galaxies are marked with green
points.  These signatures would only appear after a few megayears after
the occurrence of a quenching event.  

Investigation of local environment, gas supply for star formation, and black hole activity was done to see if these factors could have contributed to the very different star formation histories of late-type and early-type galaxies. A catalogue of galaxy halo masses and information as to whether the green valley galaxies in question were central or satellites in their clusters was used to investigate the differences in galactic environments. Schawinski et al. found that there were striking differences in environments for galaxies of both populations traveling through the green valley. Though early types were found in both low- and high-mass haloes, late types had a dramatic split. Blue cloud late types were mostly in low-mass haloes, but those in the green valley and red sequence were almost exclusively in high-mass halos. This is a possible indication that late-type quenching is caused by environmental processes. In accordance with the large percentage of late types in the green valley relative to early types, it was found that late types have large gas reservoirs relative to early types to fuel star formation and slow the evolution through the green valley. Lastly, the growth of supermassive black holes was investigated to see how these culprits of quenching may differ in early and late types. Figure 3 indicates samples in the green valley that had Active Galactic Nuclei (AGN) activity.  Since several hundred Myr or more must elapse between the end of star formation in early types and the detection of an optical AGN, it is likely that AGN are not responsible for the rapid quenching of star formation in early types and rather an after-effect of the event that triggered quenching. 

Schawinski et al. concluded with discussion on the evolutionary tracks related to the end of star formation for late and early types. Morphological classifications of SDSS images in Galaxy Zoo as well as ultraviolet photometric analysis to probe star formation histories have led to conclusions on how early- and late-type galaxies transition through the green valley. Figure 4 and figure 5 show cartoons of the predicted evolutionary sequence through the green valley for early- and late-type galaxies.

This study concluded that late-type galaxies initiate their quenching processes when they are cut off from reservoirs of cosmic gas fueling their ongoing star formation. This can happen when the galactic halo reaches a critical mass that prevents further accretion or if cooling the hot halo gas becomes inefficient. Though the star formation rate begins to decline, the stellar mass may continue to increase as the remaining gas reservoirs are converted to stars. Slowly, the galaxy moves out of the blue cloud and into the green valley, with certain physical processes possibly accelerating the gas-depletion process. Black hole accretion may appear in late types after the galaxies have been quenched. This process occurs over several gigayears.  

Figure 4. Cartoon showing the evolution of late-type galaxies from the blue cloud, through the green valley, and into the red sequence.  

As for early-type galaxies, quenching of star formation is triggered by the rapid destruction of galaxy gas reservoirs, and happens too quick to be due to gas exhaustion by star formation alone. These galaxies immediately leave the main sequence as the SFR approaches zero and stellar mass ceases to increase. As fast as stellar evolution allows, these galaxies move through the green valley and into the red sequence, typically on timescales of about 1 Gyr. Since there are very few observed blue early types, it is thought this process is initiated by a merging event of two late types and the morphology transforms as the galaxy color and SFR do. After the quenching event, visible radiation from black hole accretion can be seen, and the rapid destruction of gas reservoirs suggests the involvement of unusually strong stellar processes or AGN feedback.  

Figure 5. Cartoon showing the evolution of early-type galaxies from the blue cloud, through the green valley, and into the red sequence.