Green valley galaxies in the cosmic web: internal versus environmental quenching
MMNRAS , 1–19 (2020) Preprint 8 January 2021 Compiled using MNRAS L A TEX style file v3.0
Green valley galaxies in the cosmic web: internal versusenvironmental quenching
Apashanka Das ? , Biswajit Pandey † and Suman Sarkar ‡ Department of Physics, Visva-Bharati University, Santiniketan, Birbhum, 731235, India
ABSTRACT
We analyze the SDSS data to classify the galaxies based on their colour using a fuzzyset-theoretic method and quantify their environments using the local dimension. Wefind that the fraction of the green galaxies does not depend on the environment and10% −
20% of the galaxies at each environment are in the green valley dependingon the stellar mass range chosen. Approximately 10% of the green galaxies at eachenvironment host an AGN. Combining data from the Galaxy Zoo, we find that ∼
95% of the green galaxies are spirals and ∼
5% are ellipticals at each environment.Only ∼
8% of green galaxies exhibit signs of interactions and mergers, ∼
1% havedominant bulge, and ∼
6% host a bar. We show that the stellar mass distributionsfor the red and green galaxies are quite similar at each environment. Our analysissuggests that the majority of the green galaxies must curtail their star formationusing physical mechanism(s) other than interactions, mergers, and those driven bybulge, bar and AGN activity. We speculate that these are the massive galaxies thathave grown only via smooth accretion and suppressed the star formation primarilythrough mass driven quenching. Using a Kolmogorov-Smirnov test, we do not find anystatistically significant difference between the properties of green galaxies in differentenvironments. We conclude that the environmental factors play a minor role and theinternal processes play the dominant role in quenching star formation in the greenvalley galaxies.
Key words: methods: statistical - data analysis - galaxies: formation - evolution -cosmology: large scale structure of the Universe.
One of the most coveted goals of cosmology is understand-ing the formation and evolution of galaxies in the Uni-verse. Galaxy surveys (e.g. 2DFGRS Colless et al. (2001);SDSS Strauss et al. (2002)) over the last few decades havemeasured the properties of millions of galaxies to unprece-dented accuracy. The observed galaxy properties such ascolour (Strateva et al. 2001; Hogg et al. 2003; Balogh et al.2004; Baldry et al. 2004), star formation rate and stellarage (Kauffmann et al. 2003a), bulge to disc ratio (Kauff-mann et al. 2003a), gas to stellar mass ratio (Kannappan2004) show a bimodal character. The galaxy colour is rela-tively easy to calculate. It characterizes the stellar popula-tion of a galaxy. The distribution of the optical colour showtwo distinct peaks corresponding to two galaxy populationswhich are often referred to as the ‘red sequence’ and the ‘blue ? [email protected] † [email protected] ‡ [email protected] cloud’. The two population show significant overlap and therelatively shallow and flat region between the two peaks istermed as ‘green valley’ (Wyder et al. 2007). The galaxiesin the blue cloud are predominantly the star-forming popu-lation with disk-like morphology, younger stellar populationand lower stellar mass whereas the galaxies in the red se-quence are primarily the quiescent galaxies with a dominantbulge component, older stellar population and higher stellarmass (Strateva et al. 2001; Kauffmann et al. 2003a; Blan-ton et al. 2003; Baldry et al. 2004). The morphology of agalaxy correlates well with its stellar population. However,morphological transformations alone, can not explain thebimodal distribution of galaxy colour. Studies with GalaxyZoo (Lintott et al. 2008) reveal that a significant numberof ellipticals are contained in the blue cloud (Schawinski etal. 2009) and a large number of spirals are found in the redsequence (Masters et al. 2010).Understanding the origin of the bimodal nature of thecolour distribution and its evolution holds the importantkeys to galaxy formation and evolution. Different semi-analytic models have been used to reproduce the observed © 2020 The Authors a r X i v : . [ a s t r o - ph . GA ] J a n Das, A., Pandey, B. and Sarkar, S. bimodal distribution of galaxy colour (Menci, et al. 2005;Driver, et al. 2006; Cattaneo, et al. 2006, 2007; Cameron, etal. 2009; Trayford et al. 2016; Nelson, et al. 2018; Correa,Schaye & Trayford 2019) and gain insight into the physicalprocesses leading to bimodality. Observations suggest thatthe bimodality in galaxy colour exists out to a redshift of z = 1 − z ∼ z = 1 to z = 0(Madau et al. 1996). These observations point towards a sig-nificant evolution in the galaxy properties in the recent past,which may have played an important role in the observedbimodality of their distributions.The current paradigm of galaxy formation owes its rootto a few seminal papers from the seventies (Rees & Ostriker1977; Silk 1977; White & Rees 1978; Fall & Efstathiou 1980).The baryons are believed to be shock heated to the halovirial temperature after they fall into the dark matter po-tential well. The baryons in the inner region of the halo thenradiatively cool and settle down on a dynamical time scaleto form rotationally supported disk galaxies. These galaxiesinside the dark matter halos then evolve either in isolationor through the interactions with their environment. Boththe nature (secular) versus nurture (environment) scenarioof galaxy evolution, are capable of producing observed bi-modality in the galaxy colour distribution.The secular evolution of a galaxy is not limited to mereaging of its stellar population and there are many inter-nal physical processes (e.g. mass quenching, morphologicalquenching, bar quenching) which can play a significant rolein the reduction of star formation in a disk galaxy. Kauff-mann et al. (2003a) find that the galaxies with a mass lessthan 3 × M (cid:12) show active star formation, lower surfacemass density and disk-like morphology whereas those withmasses larger than this critical value are generally quiescentgalaxies with bulge dominated morphology and higher sur-face mass density. This critical stellar mass corresponds to acritical halo mass of 10 M (cid:12) which can be associated withthe observed bimodality. The theoretical analysis (Binney2004) and studies with hydrodynamical simulations (Birn-boim & Dekel 2003; Dekel & Birnboim 2006; Kereš et al.2005; Gabor et al. 2010; Gabor & Davé 2015) suggest thatthis critical mass is associated with a transition from ‘coldmode’ to ‘hot mode’ of accretion. Simulations show thatthe cooling time is shorter than the dynamical time in ha-los with mass less than 10 M (cid:12) and they can accrete coldgas through quasi-spherical filamentary inflows maintainingtheir star formation. But in the massive halos, cooling time isway longer where a stable shock expands to the virial radiusproducing a hot medium at the virial temperature, whichprevents any cold streams from the inter-galactic mediumto penetrate through the hot gas without getting heated. Sothe virial shock heating of the halo gas in massive halos withmass greater than 10 M (cid:12) can suppress the supply of cold gas required for star formation. This quenching of star for-mation in high mass halos is known as the ‘mass quenching’.Simulations suggest that the mass quenching alone can notmaintain long-term shutdown of star formation as the shockheated hot gas eventually cool down via radiation and thencollapse at the centres of these halos to form stars (Birn-boim, Dekel, & Neistein 2007). An additional heating sourcesuch as radio mode AGN feedback can prevent such coolingand maintain the high temperature of the halo gas (Crotonet al. 2006; Bower et al. 2006; Somerville et al. 2008). Thegravitational heating due to clumpy accretion at the cen-tre of the halos (Birnboim, Dekel, & Neistein 2007; Dekel& Birnboim 2008; Dekel, Sari, & Ceverino 2009) can alsoprevent cooling and star formation. So a coupling of thevirial shock heating with these additional energy feedbackprocesses can shutdown the star formation in dark matterhalos above the critical mass.The ‘bar quenching’ and ‘morphological quenching’ aresome of the other internal processes which can shut downthe star formation in a galaxy. The presence of stellarbars in disk galaxies can suppress star formation (Hay-wood et al. 2016; Spinoso et al. 2017; James & Percival2018; George, Subramanian, & Paul 2019). The bar-inducedtorque can transfer gas from the outskirts to the centre ofthe galaxy leading to the buildup of a bulge (Combes &Sanders 1981; Debattista et al. 2004; Kormendy & Kenni-cutt 2004; Athanassoula, Machado, & Rodionov 2013). Thecentral bulge produces a deeper potential well which canstabilize the disk against collapse (Martig et al. 2009). Thepresence of the bulge is also known to trigger AGN activity(Bruce et al. 2016) and nuclear star formation turning theinner kilopersec region devoid of any cold gas (Combes &Gerin 1985; Fang et al. 2013; Spinoso et al. 2017).Besides these internal processes, the environment of agalaxy can also play a significant role in the quenching ofstar formation. Major mergers of galaxies can transform spi-rals to ellipticals (Toomre & Toomre 1972; Barnes 2002).They can also facilitate the ejection of interstellar mediumthrough starburst and AGN or shock-driven winds (Cox etal. 2004; Murray, Quataert, & Thompson 2005; Springel, DiMatteo, & Hernquist 2005) thereby quenching star forma-tion in those galaxies. Some of the other important routesfor environmental quenching are the ram pressure strip-ping of cold gas (Gunn & Gott 1972), galaxy harassmentin clusters (Moore et al. 1996; Moore, Lake, & Katz 1998),strangulation in galaxy-group interactions (Gunn & Gott1972; Balogh, Navarro, & Morris 2000) and starvation due totruncation of gas supply (Larson, Tinsley, & Caldwell 1980;Somerville & Primack 1999; Kawata & Mulchaey 2008).The blue galaxies evolve to red galaxies via quenchingof star formation. The galaxies with intermediate proper-ties between blue and red galaxies lie in the green valleyundergoing such a transition. It is important to understandthe primary physical mechanisms responsible for such tran-sition. The red galaxies are preferably found in the denserregion (Hoyle et al. 2002; Park et al. 2005) which suggeststhe role of environment in quenching. Some studies suggestthat AGN feedback plays an important role in quenchingstar formation in the transitional green valley population(Nandra et al. 2007; Hasinger 2008; Silverman et al. 2008;Cimatti et al. 2013). The analysis of Galaxy Zoo by Schaw-inski et al. (2014) suggest that quenching in green valley MNRAS , 1–19 (2020) reen valley in the cosmic web galaxies require the gas reservoir destruction caused by star-burst and AGN feedback in major mergers. Starving galax-ies of the cold gas supply and exhaustion of the remaininggas can also initiate such quenching. Lin et al. (2017) findlower star formation efficiency in the green valley galaxiesfrom the study of their cold molecular gas content, whichindicates that complete removal of cold gas supply is notnecessary for quenching in these galaxies. A comprehensivereview of the green valley galaxies can be found in Salim(2014). Coenda, Martínez, & Muriel (2018) study the prop-erties of green valley galaxies in fields, groups and clustersusing SDSS and find that there is a clear environmentaldependence of external quenching mechanisms in the greenvalley galaxies. Jian et al. (2020) study the redshift evo-lution of green valley galaxies in different environments tofind a mild evolution in the environmental dependence whichsuggests that slow quenching mechanisms are operating indenser environments since z ∼ ∼ h − Mpc.Using SDSS, Sarkar & Pandey (2020) show that the mu-tual information between morphology and large-scale en-vironment are statistically significant at 99 .
9% confidencelevel. A recent analysis by Bhattacharjee, Pandey, & Sarkar (2020) show that conditioning on stellar mass does not pro-vide a complete explanation of the mutual information be-tween morphology and the large-scale environment. Pandey& Sarkar (2020) analyze the data from SDSS to find thatat a fixed density, the fraction of red and blue galaxies aresensitive to the geometric environments of the cosmic web.The SDSS (Strauss et al. 2002) is the most successfulredshift surveys to date. It provides the photometric andspectroscopic information of millions of galaxies which al-lows us to carry out statistical analysis of the data and ad-dress many important issues related to galaxy formation andevolution. The galaxies in the green valley hold importantclues about galaxy evolution. It is important to understandthe role of internal and external influences on the suppres-sion of star formation in the green valley galaxies. In thepresent work, we plan to study the fraction of green galax-ies and their possible routes of quenching in different geo-metric environments of the cosmic web. As mass plays animportant role in quenching, the environmental dependenceof the fractions should be investigated at fixed stellar mass.Analysis in different mass bins may help us to identify theinfluence of the environments on quenching. It is also wellknown that the presence of a dominant bulge, a bar or AGNactivity can quench star formation in a galaxy (Fisher &Drory 2008; Fabian 2012; Lang et al. 2014; Förster Schreiberet al. 2014; Leslie et al. 2016; Haywood et al. 2016; George,Subramanian, & Paul 2019). We would like to investigatethe relative importance of these quenching agents in differ-ent environments of the cosmic web. It is also important tocompare the distributions of the stellar mass, star formationrate and star formation history of the green valley galaxiesin different geometric environments to understand the roleof environments in deciding the properties of green galaxies.We also address if the length scales associated with the hostenvironments affect the galaxies in the green valley.The paper is organised as follows, we describe the datain Section 2, the method of analysis in Section 3, discuss theresults in Section 4 and present our conclusions in Section5.
The Sloan Digital Sky Survey (SDSS) is a multi-band pho-tometric and spectroscopic redshift survey that uses a dedi-cated 2.5 m telescope at Apache Point Observatory in NewMexico to measure the spectra and images of millions ofgalaxies over roughly one third of the sky. Gunn et al. (1998)describe the technical details of the SDSS photometric cam-era. The construction, design and performance of the SDSStelescope is described in Gunn et al. (2006). The algorithmfor selecting the SDSS main sample for spectroscopy is pro-vided in Strauss et al. (2002). A technical summary of theSDSS is outlined in York et al. (2000).We use the publicly available data from SDSS DR16(Ahumada et al. 2020) for the present analysis. We re-trieve the required data from
Casjobs using StructuredQuery Language . We have downloaded the spectroscopicand photometric information of galaxies from
SpecPhotoAll https://skyserver.sdss.org/casjobs/MNRAS000
SpecPhotoAll https://skyserver.sdss.org/casjobs/MNRAS000 , 1–19 (2020) Das, A., Pandey, B. and Sarkar, S. table using the following cuts: z < .
2, 13 . ≤ r p < . ◦ ≤ α ≤ ◦ , 0 ◦ ≤ δ ≤ ◦ where z is the redshift, r p isthe r -band Petrosian magnitude, α and δ are right ascen-sion and declination respectively. We also download stel-lar mass and star formation rate of these galaxies from stellarMassF SP SGranW ideNoDust table (Conroy et al.2009), K correction from P hotoz table and 4000Å breakstrengh D4000 (Bruzual 1983) from galSpecIndx table(Brinchmann et al. 2004). We use galSpecExtra table de-rived from MPA-JHU spectroscopic catalogue (Brinchmannet al. 2004; Kauffmann et al. 2003b) to identify the AGNsclassified on the basis of BPT emission line diagram. Thegalaxies with AGN activity are flagged as 4 in this table.For morphological classification we use zooSpec table de-rived from Galaxy Zoo (Lintott et al. 2008). We identify theelliptical and spiral galaxies from this table as those whichhave their elliptical and spiral flag set to 1 (with debiasedvote fraction > .
8) respectively. We combine these tablesand apply the above mentioned cuts which provides us with321741 galaxies. We construct a volume limited sample fromthis data by restricting the K-corrected and extinction cor-rected r -band absolute magnitude to − ≥ M r ≥ −
23 whichcorresponds to a redshift cut of 0 . ≤ z ≤ . m = 0 . , Ω Λ0 = 0 .
685 and h = 0 .
674 (Planck Col-laboration et al. 2020).
We classify the galaxies as red, blue and green using a fuzzyset theory based method proposed in Pandey (2020). A fuzzyset A is defined by a set of ordered pairs (Zadeh 1965), A = (cid:8) ( x, µ A ( x ) ) | x ∈ X (cid:9) (1)where X is the Universal set and µ A ( x ) is the membershipfunction of the fuzzy set A . The membership function mapsthe elements of the Universal set X to real numbers in [0 , A . In other words, it measures the degree of membership ofa particular element in the fuzzy set.We define the fuzzy set R corresponding to redness ofall galaxies in the volume limited sample as, R = (cid:8) ( u − r, µ R ( u − r )) | ( u − r ) ∈ X (cid:9) (2)where X is the Universal set of ( u − r ) colour. The member-ship to the fuzzy set is described with a sigmoidal function(Pandey 2020), µ R ( u − r ; a, c ) = 11 + e − a [( u − r ) − c ] (3)Here a and c are constants. We choose a = 5 . c = 2 . u − r colour distribution. The two peaks of the bimodal dis-tribution corresponds to the ‘blue cloud’ and ‘red sequence’.The parameters c and a respectively denote the crossoverpoint of the fuzzy set R and the slope at the crossover point.The fuzzy set has the maximum uncertainty at the crossover point and the choice of c = 2 . u − r ) ∼ .
2. The value of a is chosen so as to ensure that the galaxy with largest andsmallest ( u − r ) colour respectively have their membershipfunction 1 and 0 in the fuzzy set R . Once the fuzzy setfor ‘redness’ is defined, the fuzzy set B for ‘blueness’ can besimply obtained by taking a fuzzy complement of set R . Themembership function µ B ( u − r ) of the fuzzy set B is definedas, µ B ( u − r ) = 1 − µ R ( u − r ) , ∀ ( u − r ) ∈ X (4)The fuzzy set G for ‘greenness’ can be defined by takinga fuzzy intersection of R and B . The membership function µ G ( u − r ) of the fuzzy set G is defined as, µ G ( u − r ) = 2 min (cid:8) µ R ( u − r ) , µ B ( u − r ) (cid:9) , ∀ ( u − r ) ∈ X (5)Here min denotes the minimum operator. The fuzzy set G has the maximum height of 0 . u − r ) = 2 . µ G domi-nates µ B and µ R . The red and blue galaxies are also definedin a similar manner. In the present analysis, we find that thegalaxies with 2 . < ( u − r ) < .
33 are green, ( u − r ) ≤ . u − r ) ≥ .
33 are red (Figure 1). These cuts pro-vide us with 51965 red, 27967 blue and 13255 green galaxies.To study the detailed morphology of green galax-ies, we further cross match their specObjid using the zoo MainSpecz table derived from Galaxy Zoo 2 (Willett etal. 2013). We obtain the information regarding the presenceof bulge, bar, disturbed, irregular or merger features in thegreen galaxies in our sample. We only consider the classifica-tions with a debiased vote fraction greater than 0 . The galaxies reside in different types of geometric environ-ments in the cosmic web. We quantify the geometric envi-ronment of a galaxy using the local dimension (Sarkar &Bharadwaj 2009). The local dimension of a galaxy is simplybased on the number counts of galaxies within a sphere ofradius R centered around it. The number count N ( < R ) isexpected to scale as, N ( < R ) = AR D (6)where D is the local dimension and A is a constant. We varythe radius of the measuring sphere over a range of lengthscales R h − Mpc ≤ R ≤ R h − Mpc. All the galaxies forwhich the radius of the measuring sphere can be varied inthis range and there are at least 10 galaxies within the twoconcentric spheres of radius R and R , are included in theanalysis. For each valid centers, we fit the measured numbercounts N ( < R ) within R and R to Equation 6. The bestfit values of A and D are determined using a least-squarefitting. We also estimate the associated χ value using thefitted and observed values of N ( < R ). We further restrict ouranalysis to only those centers for which chi-square per degree MNRAS , 1–19 (2020) reen valley in the cosmic web Figure 1.
This figure shows the definition of the red, blue and green galaxies classified using fuzzy set theory. The region between thetwo vertical solid lines corresponds to the green galaxies. of freedom χ ν ≤ . R = 2 h − Mpc throughout the present analysis. We havechosen R = 10 h − Mpc and R = 40 h − Mpc to probe thegeometric environments of galaxies on two different lengthscales. The geometric environment around a galaxy withinlength scale range R h − Mpc ≤ R ≤ R h − Mpc is charac-terized by its local dimension D . The galaxies residing atthe center of a straight filament are expected to have a localdimension of D = 1. A local dimension of D = 2 would repre-sent the galaxies lying within a sheet-like structure whereasthe galaxies with D = 3 are expected to be distributed ho-mogeneously in a three dimensional volume. The cosmic webis an interconnected complex network of different morpho-logical components which vary widely in their shapes andsizes. Very often, the measuring sphere may include multi-ple types of morphological components (e.g. a filament anda sheet) leading to intermediate values of local dimension D . We classify the galaxies belonging to different types ofgeometric environments by assigning a specific range of lo-cal dimension to each class (Table 1). In this classification,the D D D D D D D . D . R = 10 h − Mpc whereas at 40 h − Mpc, the local dimen-sion can be computed for 16480 red, 9341 blue and 4418green galaxies.
If environment plays a significant role in the transition ofgreen valley galaxies then the distributions of their proper-ties should not be same in different environments. For anygiven property of green galaxy, we compare the cumulativedistributions in two different environments with the two-sample Kolmogorov-Smirnov (KS) test. The KS test doesnot make any assumptions about the distributions. The nullhypothesis associated with the test assumes that the chosenproperty of the green galaxies from different environmentsare actually sampled from identical distributions. We definethe supremum difference D KS between the two cumulativedistribution functions as, D KS = sup X { | f ,m ( X ) − f ,m ( X ) | } (7)where f ,m ( X ) and f ,m ( X ) are the cumulative distributionfunctions of the chosen property ( X ) of the green galaxies atthe m th bin in type-1 and type-2 environments respectively.Here the properties that we have chosen are the stellar mass,star formation rate and stellar age. The type-1 and type-2 MNRAS000
If environment plays a significant role in the transition ofgreen valley galaxies then the distributions of their proper-ties should not be same in different environments. For anygiven property of green galaxy, we compare the cumulativedistributions in two different environments with the two-sample Kolmogorov-Smirnov (KS) test. The KS test doesnot make any assumptions about the distributions. The nullhypothesis associated with the test assumes that the chosenproperty of the green galaxies from different environmentsare actually sampled from identical distributions. We definethe supremum difference D KS between the two cumulativedistribution functions as, D KS = sup X { | f ,m ( X ) − f ,m ( X ) | } (7)where f ,m ( X ) and f ,m ( X ) are the cumulative distributionfunctions of the chosen property ( X ) of the green galaxies atthe m th bin in type-1 and type-2 environments respectively.Here the properties that we have chosen are the stellar mass,star formation rate and stellar age. The type-1 and type-2 MNRAS000 , 1–19 (2020)
Das, A., Pandey, B. and Sarkar, S.
D1 D1.5 D2 D2.5 D3050001000015000200002500030000 N u m be r R = 10 h -1 Mpc
AllGreenBlueRed
D1 D1.5 D2 D2.5 D3050001000015000 N u m be r R = 40 h -1 Mpc
AllGreenBlueRed
D1 D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 10 h -1 Mpc
GreenBlueRed
D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 40 h -1 Mpc
GreenBlueRed
Figure 2.
The top left and top right panel shows the number of red, blue and green galaxies in different environments for R = 10 h − Mpcand R = 40 h − Mpc. The respective fractions are shown in the bottom left and bottom right panels. The 1 − σ errorbar shown here areestimated using 10 jackknife samples drawn from the volume limited sample. can be any environments out of D D . D D . D m ∈ { , , ...., N } and sup denotes the supremumof all the ( N + N ) differences such that P N m =1 f ,m ( X ) = P N m =1 f ,m ( X ) = 1.The critical value of the supremum difference associatedwith a given significance level ( α ) is given by, D KS ( α ) = s − ln (cid:16) α (cid:17) N + N N N (8)Here N and N are the number of green galaxies intype-1 and type-2 environments respectively. When D KS >D KS ( α ), we can reject the null hypothesis at a significancelevel α . The null hypothesis can be tested at different signif-icance level to find if the distribution of a given property ofgreen galaxies are significantly different in the two separateenvironments considered. We show the number of red, blue and green galaxies avail-able in different environments of the cosmic web in the toptwo panels of Figure 2. The top left and right panel ofthis figure respectively correspond to R = 10 h − Mpc and R = 40 h − Mpc. The top left panel shows that the numberof galaxies for each class peaks at local dimension of D = 1 . h − Mpc have a morphology which is interme-diate between filaments ( D = 1) and sheets ( D = 2). Thesestructures could represent the curved filaments or the inter-secting regions between filaments and sheets. The numberof galaxies in different classes peak at the local dimension of D = 2 in the top right panel which indicates that the cosmicweb is dominated by the sheet-like patterns when analyzedon a length scale of 40 h − Mpc. There is a clear shift ofthe location of the peak with increasing length scales whichsuggests that there is a gradual transition of the local dimen-sion of the morphological patterns in the cosmic web withincreasing length scales. The fraction of red, blue and greengalaxies in different environments for R = 10 h − Mpc and R = 40 h − Mpc are respectively shown in the two bottom
MNRAS , 1–19 (2020) reen valley in the cosmic web D1 D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 10 h -1 Mpc10 ≤ log(M stellar /M sun ) ≤ RedBlueGreen
D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 40 h -1 Mpc10.5 ≤ log(M stellar /M sun ) ≤ RedBlueGreen
D1 D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 10 h -1 Mpc10.75 ≤ log(M stellar /M sun ) ≤ RedBlueGreen
D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 40 h -1 Mpc11 ≤ log(M stellar /M sun ) ≤ RedBlueGreen
Figure 3.
The top left and bottom left panels show the fractions of red, blue and green galaxies in different environments for R =10 h − Mpc. These correspond to two different mass bins as mentioned in the panels. The results for two different mass bins for R =40 h − Mpc are shown in the top right and bottom right panels of the figure. The 1 − σ errorbar shown are estimated using 10 jackknifesamples drawn from the volume limited sample. Table 1.
This table defines the different geometric environments in the cosmic web based on the local dimension of a galaxy.Local dimension : 0 . ≤ D < .
25 1 . ≤ D < .
75 1 . ≤ D < .
25 2 . ≤ D < . D ≥ . D D . D D . D panels of Figure 2. The results suggest that the fraction ofred galaxies decreases with the increasing local dimensionof their host environment. We see a reverse trend for theblue galaxies. These trends were reported earlier in a paperby Pandey & Sarkar (2020). In the present work, our pri-mary focus is to understand the abundance of green galaxiesin different environments of the cosmic web and the phys-ical processes that drives the transition in these galaxies.We also show the fraction of green galaxies in different ge-ometric environments of the cosmic web in the two bottompanels of Figure 2. We find that ∼
15% galaxies in each geo-metric environment are green. The fraction of green galaxiesare nearly independent of the local dimension of their hostenvironments.Mass of a galaxy plays an important role in quenchingstar formation. The star formation is quenched in galax-ies above a critical mass. The massive galaxies are morecommon in high density environments. It is known that the geometric environments with smaller local dimension havehigher local densities (Pandey & Sarkar 2020). In general,filaments are denser than sheets and the sheets are denserthan the fields. So a higher fraction of red galaxies in fila-ments than sheets and fields may simply arise due to a higherabundance of more massive galaxies in filaments comparedto the other geometric environments. A similar argumentapplies for the blue galaxies which have a relatively lowermass and are more abundant in low density environmentswith higher local dimension. Thus a higher fraction of galax-ies are expected to be mass quenched in environments withsmaller local dimension. The green galaxies are intermedi-ate between red and blue galaxies and it is interesting to seethat the fraction of green galaxies is nearly independent ofenvironment in Figure 2.
MNRAS000
MNRAS000 , 1–19 (2020)
Das, A., Pandey, B. and Sarkar, S.
D1 D1.5 D2 D2.5 D3010002000300040005000 N u m be r R = 10 h -1 Mpc
GreenGreen AGN
D1 D1.5 D2 D2.5 D305001000150020002500 N u m be r R = 40 h -1 Mpc
GreenGreen AGN
D1 D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 10 h -1 Mpc
AGN
D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 40 h -1 Mpc
AGN
Figure 4.
The top left and top right panels show the number of green galaxies and green AGNs in different environments for R = 10 h − Mpc and R = 40 h − Mpc respectively. The bottom left and right panels shows the respective fractions of AGNs at eachenvironment. The 1 − σ errorbar shown are estimated using 10 jackknife samples drawn from the green galaxy sample. In order to understand the role of mass quenching in differ-ent environments, we divide the red, blue and green galax-ies into two different mass bins and study their fractionsin each environment. The top left and bottom left panelsof Figure 3 show the results in two different mass bins for R = 10 h − Mpc. The top left panel of Figure 3 shows thatat lower masses, a higher fraction of blue galaxies are presentat each environment. This is opposite to what is observedin Figure 2. Here the fraction of blue galaxies also increaseswith increasing local dimension as noted earlier in Figure 2.The fraction of red galaxies are smaller than blue galax-ies at each environment and the red fraction decreases withthe increasing local dimension as before. Only ∼
10% galax-ies at each environment are green and the fraction of greengalaxies is independent of environment in the lower massbin. Now we shift our attention to the bottom left panelwhere the fractions are shown for the galaxies in the highermass bin. We note that at higher mass bin, each environ-ment is dominated by the red galaxies. The fractions of redgalaxies are now significantly higher (2 − ∼
18% in this case) also increases in the higher mass bin which again remains nearly independent of envi-ronment. We observe a change in the green fraction in thiscase due to the fact that all the galaxies in the higher massbin have masses above the critical mass (3 × M (cid:12) ) re-quired for mass quenching whereas a significant number ofgalaxies in the lower mass bin have masses below this criti-cal value. The result suggest that mass quenching may playan important role in suppressing star formation in the greenvalley galaxies.The corresponding results for R = 40 h − Mpc areshown in the top right and bottom right panels of Figure 3.We see that the red galaxies dominate each environment inboth the mass bins. Noticeably at higher masses, the fractionof blue galaxies show a decrease in each environment. Thedecrease is most pronounced for the field galaxies ( D −
20% inall the environments. Only a mild increase in the fraction ofgreen galaxies are observed in the higher mass bin. It may
MNRAS , 1–19 (2020) reen valley in the cosmic web D1 D1.5 D2 D2.5 D3010002000300040005000 N u m be r R = 10 h -1 Mpc
GreenGreen uncertainGreen ellipticalGreen spiral
D1 D1.5 D2 D2.5 D305001000150020002500 N u m be r R = 40 h -1 Mpc
GreenGreen uncertainGreen ellipticalGreen spiral
D1 D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 10 h -1 Mpc
EllipticalSpiral
D1.5 D2 D2.5 D300.20.40.60.81 F r a c t i on R = 40 h -1 Mpc
EllipticalSpiral
Figure 5.
The top left and top right panel show the number of green galaxies with elliptical, spiral and uncertain morphology indifferent environments for R = 10 h − Mpc and R = 40 h − Mpc respectively. The bottom left and bottom right panels show therespective fractions of classified green galaxies with spiral and elliptical morphology. The 1 − σ errorbar shown are estimated using 10jackknife samples drawn from the sample of green galaxies. D1 D1.5 D2 D2.5 D300.050.10.15 F r a c t i on R = 10 h -1 Mpc
Dominant bulgeBarDisturbedIrregularMerger
D1.5 D2 D2.5 D300.050.10.15 F r a c t i on R = 40 h -1 Mpc
Dominant bulgeBarDisturbedIrregularMerger
Figure 6.
The left and right panels separately show the fraction of green galaxies with the dominant bulge, bar and those with thedisturbed, irregular and merger features in different environments for R = 10 h − Mpc and R = 40 h − Mpc.MNRAS000
The left and right panels separately show the fraction of green galaxies with the dominant bulge, bar and those with thedisturbed, irregular and merger features in different environments for R = 10 h − Mpc and R = 40 h − Mpc.MNRAS000 , 1–19 (2020) Das, A., Pandey, B. and Sarkar, S. log(M stellar /M sun ) pd f R = 10 h -1 MpcD1
Green - 2080Red - 8388Blue - 4393 log(M stellar /M sun ) pd f R = 40 h -1 MpcD1.5
Green - 167Red - 893Blue - 252 log(M stellar /M sun ) pd f R = 10 h -1 MpcD2
Green - 2786Red - 10698Blue - 5996 log(M stellar /M sun ) pd f R = 40 h -1 MpcD2
Green - 2011Red - 8690Blue - 3989 log(M stellar /M sun ) pd f R = 10 h -1 MpcD3
Green - 195Red - 564Blue - 414 log(M stellar /M sun ) pd f R = 40 h -1 MpcD3
Green - 320Red - 812Blue - 896
Figure 7.
The top left, middle left and bottom left panels show the stellar mass distributions of red, blue and green galaxies in,respectively D D D R = 10 h − Mpc. The same for R = 40 h − Mpc are shown in the top right, middleright and bottom right panels of the figure. The number of red, blue and green galaxies available in each case are mentioned in therespective panels. be noted that the stellar mass of each galaxies in both themass bins are greater than the critical mass.The sensitivity of the red and blue fractions to the ge-ometry and density of their environment may not solely arisedue to the mass quenching. External influences of their en-vironment may also play a significant role in quenching starformation in these galaxies. However it is worth noting that the fraction of green galaxies are nearly independent of theirenvironments.
MNRAS , 1–19 (2020) reen valley in the cosmic web log(M stellar /M sun ) pd f R = 10 h -1 Mpc
D1 - 2080D1.5 - 4211D2 - 2786D2.5 - 908D3 - 195 log(M stellar /M sun ) pd f R = 40 h -1 Mpc
D1.5 - 167D2 - 2011D2.5 - 1920D3 - 320
10 10.5 11 11.5 12 log(M stellar /M sun ) c d f R = 10 h -1 Mpc
Green D1 - 2080Green D2 - 2786
10 10.5 11 11.5 12 log(M stellar /M sun ) c d f R = 10 h -1 Mpc
Green D2 - 2786Green D3 - 195
10 10.5 11 11.5 12 log(M stellar /M sun ) c d f R = 10 h -1 Mpc
Green D1 - 2080Green D3 - 195
10 10.5 11 11.5 12 log(M stellar /M sun ) c d f R = 40 h -1 Mpc
Green D2 - 2011Green D3 - 320
Figure 8.
The top left and top right panels compare the stellar mass distribution of green galaxies in different environments for R = 10 h − Mpc and R = 40 h − Mpc respectively. The number of green galaxies available at different environments are mentioned inthe top two panels. The middle and bottom panels compare the CDF of the stellar mass distribution for the green galaxies. The type ofenvironments between which the comparison is made and the associated length scales are mentioned in each panel.
AGN feedback is believed to play an important role inquenching star formation in galaxies. It prevents cooling ofhot halo gas and supports long term shutdown of star forma-tion in massive galaxies. We explore the fraction of AGN ingreen galaxies in different environments of the cosmic web. The top left and right panels of Figure 4 show the numberof green galaxies and green AGN in different environmentsfor R = 10 h − Mpc and R = 40 h − Mpc respectively. Thefractions of green AGNs in different environments for thesetwo length scales are shown in the two bottom panels of Fig-ure 4. We find that ∼
10% green galaxies host AGN in eachenvironment. The fraction of AGN in green galaxies are in-
MNRAS000
MNRAS000 , 1–19 (2020) Das, A., Pandey, B. and Sarkar, S.
SFR ( M sun yr -1 ) pd f R = 10 h -1 MpcD1
Blue - 4393Red - 8388Green - 2080
SFR ( M sun yr -1 ) pd f R = 40 h -1 MpcD1.5
Blue - 252Red - 893Green - 167
SFR ( M sun yr -1 ) pd f R = 10 h -1 MpcD2
Blue - 5996Red - 10698Green - 2786
SFR ( M sun yr -1 ) pd f R = 40 h -1 MpcD2
Blue - 3989Red - 8690Green - 2011
SFR ( M sun yr -1 ) pd f R = 10 h -1 MpcD3
Blue - 414Red - 564Green - 195
SFR ( M sun yr -1 ) pd f R = 40 h -1 MpcD3
Blue - 896Red - 812Green - 320
Figure 9.
Same as Figure 7 but for star formation rate. dependent of environments and the associated length scales.This indicates that the AGN activity in green galaxies couldbe a result of secular evolution. At each environment, only asmall fraction of green galaxies host AGN. The destructionof the gas reservoirs in galaxies by AGN-driven winds pro-vide an effective channel for quenching. However, the tran-sition in majority of green galaxies can not be explained bythe AGN driven quenching of star formation.
The morphology and the star formation activity of a galaxyare known to be closely correlated with each other. In or-der to understand the role of morphological quenching in thegreen valley galaxies, we study their morphology as classifiedvisually by a large number of volunteers in the Galaxy Zoo(Lintott et al. 2008) and the Galaxy Zoo 2 project (Willett etal. 2013). We find that out of 13255 green galaxies there are
MNRAS , 1–19 (2020) reen valley in the cosmic web SFR ( M sun yr -1 ) pd f R = 10 h -1 Mpc
D1 - 2080D1.5 - 4211D2 - 2786D2.5 - 908D3 - 195
SFR ( M sun yr -1 ) pd f R = 40 h -1 Mpc
D1.5 - 167D2 - 2011D2.5 - 1920D3 - 320
SFR ( M sun yr -1 ) c d f R = 10 h -1 Mpc
Green D1 - 2080Green D2 - 2786
SFR ( M sun yr -1 ) c d f R = 10 h -1 Mpc
Green D2 - 2786Green D3 - 195
SFR ( M sun yr -1 ) c d f R = 10 h -1 Mpc
Green D1 - 2080Green D3 - 195
SFR ( M sun yr -1 ) c d f R = 40 h -1 Mpc
Green D2 - 2011Green D3 - 320
Figure 10.
Same as Figure 8 but for star formation rate.
370 ellipticals , 6864 spirals and 6021 galaxies with uncertainmorphology. We first show the number of spirals, ellipticalsand galaxies with uncertain morphology in different envi-ronments for R = 10 h − Mpc and R = 40 h − Mpc in thetop two panels of Figure 5. We plot the respective fractionsof spirals and ellipticals in different environments in the twobottom panels of this figure. The results show that ∼ ∼
5% galaxiesare ellipticals in each environments. These 5% ellipticals in the green valley could be the passive galaxies which mayhave underwent a wet merger causing a rejuvenation of starformation (Kaviraj et al. 2009; Thomas et al. 2010). Themajority of the classified green galaxies are spirals. We ob-tain the detailed morphology of a subset of the green galax-ies (9198 out of 13255) from Galaxy Zoo 2. We find thatout of these 9198 green galaxies, there are 71 galaxies withdominant bulge, 586 galaxies with bar, 353 galaxies withdisturbed features, 253 galaxies with irregular features and
MNRAS000
MNRAS000 , 1–19 (2020) Das, A., Pandey, B. and Sarkar, S.
D4000 pd f R = 10 h -1 MpcD1
Green - 2080Blue - 4393Red - 8388
D4000 pd f R = 40 h -1 MpcD1.5 ,
Green - 167Blue - 252Red - 893
D4000 pd f R = 10 h -1 MpcD2
Green - 2786Blue - 5996Red - 10698
D4000 pd f R = 40 h -1 MpcD2
Green - 2011Blue - 3989Red - 8690
D4000 pd f R = 10 h -1 MpcD3
Green - 195Blue - 414Red - 564
D4000 pd f R = 40 h -1 MpcD3
Green - 320Blue - 896Red - 812
Figure 11.
Same as Figure 7 but for stellar age.
73 galaxies with merger features. In each case, we only con-sider the classifications with a debiased vote fraction greaterthan 0.8. We show the fractions of green galaxies havingdominant bulge, bar and those with disturbed, merger andirregular features in different environment in the two panelsof Figure 6. In both the panels, these fractions are nearlyindependent of environment.
In the top left, middle left and bottom left panels of Fig-ure 7, we show the distributions of the stellar mass for thered, blue and green galaxies in D D D R = 10 h − Mpc. The results for R = 40 h − Mpc in D . D D MNRAS , 1–19 (2020) reen valley in the cosmic web D4000 pd f R = 10 h -1 Mpc
D1 - 2080D1.5 - 4211D2 - 2786D2.5 - 908D3 - 195
D4000 pd f R = 40 h -1 Mpc
D1.5 - 167D2 - 2011D2.5 - 1920D3 - 320
D4000 c d f R = 10 h -1 Mpc
Green D1 - 2080Green D2 - 2786
D4000 c d f R = 10 h -1 Mpc
Green D2 - 2786Green D3 - 195
D4000 c d f R = 10 h -1 Mpc
Green D1 - 2080Green D3 - 195
D4000 c d f R = 40 h -1 Mpc
Green D2 - 2011Green D3 - 320
Figure 12.
Same as Figure 8 but for stellar age. red and green galaxies are quite similar and are noticeablydifferent than that for the blue galaxies. The stellar massdistribution of the blue galaxies extends to lower masssescompared to the red and green galaxies in each type of en-vironment.We also compare the stellar mass distributions of greengalaxies in different environments in the top two panels ofFigure 8. The left and right panels in this figure correspondto R = 10 h − Mpc and R = 40 h − Mpc respectively. Wetest the statistical significance of the differences between the distributions using a two-sample Kolmogorov-Smirnov test.The middle and bottom panels of Figure 8 compare the CDFof stellar mass distribution in different environments. The D KS values from the test listed in Table 2, suggests that at R = 10 h − Mpc, the null hypothesis can not be rejected athigh confidence level. So there are no statistically significantdifference between the stellar mass distributions of greengalaxies in different environments. However we note thatthe stellar mass distribution of green galaxies in D MNRAS000
Same as Figure 8 but for stellar age. red and green galaxies are quite similar and are noticeablydifferent than that for the blue galaxies. The stellar massdistribution of the blue galaxies extends to lower masssescompared to the red and green galaxies in each type of en-vironment.We also compare the stellar mass distributions of greengalaxies in different environments in the top two panels ofFigure 8. The left and right panels in this figure correspondto R = 10 h − Mpc and R = 40 h − Mpc respectively. Wetest the statistical significance of the differences between the distributions using a two-sample Kolmogorov-Smirnov test.The middle and bottom panels of Figure 8 compare the CDFof stellar mass distribution in different environments. The D KS values from the test listed in Table 2, suggests that at R = 10 h − Mpc, the null hypothesis can not be rejected athigh confidence level. So there are no statistically significantdifference between the stellar mass distributions of greengalaxies in different environments. However we note thatthe stellar mass distribution of green galaxies in D MNRAS000 , 1–19 (2020) Das, A., Pandey, B. and Sarkar, S. D .
9% confidence level at R = 40 h − Mpc.
We show the distributions of star formation rate (SFR) inred, blue and green galaxies in different types of environmentin Figure 9. The left and right panels of Figure 9 show the re-sults for R = 10 h − Mpc and R = 40 h − Mpc respectively.The SFR distribution of red galaxies at each environmentpeaks at the lowest SFR bin. The peak of the SFR distri-bution of green galaxies occur at the lowest SFR bin too.However, we see that ∼
80% red galaxies and ∼
30% greengalaxies reside in the lowest SFR bin. So there is a largedifference between the amplitudes of the peaks associatedwith the two distributions. The distribution of green galax-ies dominate the distribution of red galaxies in all the otherSFR bins. The SFR distribution of green galaxies extendsto higher SFR values as compared to the distribution of redgalaxies. The distribution of blue galaxies peaks at higherSFR and extends further compared to that for the red andgreen galaxies at each environment.We compare the SFR distributions of the green galax-ies in different types of environments for R = 10 h − Mpcand R = 40 h − Mpc in the top two panels of Figure 10. Wetest the statistical significance of the differences between thedistribution using a Kolmogorov-Smirnov test. The CDFs indifferent environments are compared in the middle and bot-tom panels of Figure 10 and the corresponding D KS valuesare listed in Table 2. The results show that the null hypoth-esis can not be rejected even at 80% confidence level whichimplies that the SFR in green galaxies are independent oftheir environment. We show the distributions of stellar age as characterized by D R = 10 h − Mpc.The results for R = 40 h − Mpc are shown in the top, mid-dle and bottom right panels of Figure 11. At each panel, thedistributions of the red and blue galaxies show two distinctpeaks at D . D . D KS values in Table 2. The test shows that thenull hypothesis can not be rejected even at 80% confidencelevel which indicates that the stellar age of green galaxiesare independent of their environment. In the present work, we analyze a volume limited samplefrom the SDSS to understand the internal and external in-fluences on the quenching of star formation in the greenvalley galaxies. We employ a fuzzy set-theoretic method toclassify the galaxies according to their colour and character-ize their environment using the local dimension. We studythe fraction of green galaxies in different environments as afunction of their stellar mass. We also study the morphol-ogy of green galaxies in different environments. The AGNdriven winds can quench star formation in galaxies. Theycan also play an important role in mass driven quenching.So, we study the fraction of AGNs in green galaxies in dif-ferent environments. The presence of a dominant bulge orbar can also suppress star formation in galaxies. We measurethe fraction of green galaxies with dominant bulge and barin various environments. Galaxy interaction such as stran-gulation and harassment are also known to quench star for-mation in galaxies. We study the fraction of green galaxieswith disturbed, irregular or merger features in different en-vironments. We finally compare the distributions of stellarmass, SFR and stellar age of red, blue and green galaxiesin different environments of the cosmic web. We test if thedistributions of stellar mass, SFR and stellar age of greengalaxies in various environments have any statistically sig-nificant differences. Our main results can be summarized asfollows:(i) The fraction of green galaxies is independent of geo-metric environments of the cosmic web. Around 10% − ∼
10% of green galaxies showAGN activity. The fraction of AGN in green galaxies is alsoindependent of the environment and the associated lengthscales.(iii) At each environment, ∼
95% of the visually classi-fied green galaxies are spirals and only ∼
5% are ellipticals.The morphology of the green galaxies is also independent ofthe environment and the associated length scales.(iv) Most of the visually classified green galaxies havea bulge but only ∼
1% of them at each environment, have adominant bulge. Nearly 6% −
10% of classified green galax-ies at each environment host a bar. The bulge fraction andbar fraction in green galaxies show a mild environmentaldependence with no clear trends.(v) Only ∼ ∼
3% and ∼
1% visually classified greengalaxies at each environment respectively show disturbed,irregular and merger features. The green fractions with theseodd features show a mild environmental dependence but noobvious trends are observed.
MNRAS , 1–19 (2020) reen valley in the cosmic web Table 2.
The table below shows the Kolmogorov-Smirnov statistic D KS for comparisons of log( M stellar /M sun ), SF R and D R = 10 h − Mpc and R = 40 h − Mpc. The table also lists the critical valuesof D KS ( α ) above which null hypothesis can be rejected at different confidence levels. D KS D KS ( α ) R Environment type log( M stellar /M sun ) SF R D .
9% 99 .
5% 99% 95% 90% 80%D1-D2 0.041 0.015 0.039 0.056 0.050 0.047 0.039 0.035 0.03110 h − Mpc D2-D3 0.073 0.060 0.049 0.144 0.128 0.120 0.100 0.090 0.079D1-D3 0.075 0.051 0.055 0.146 0.129 0.121 0.101 0.091 0.08040 h − Mpc D2-D3 0.126 0.041 0.072 0.117 0.104 0.098 0.081 0.073 0.064 (vi) The stellar mass distributions of red and greengalaxies are quite similar in each environment. The distribu-tions of SFR and stellar age of red, blue and green galaxiesare noticeably different in each environment.(vii) The distributions of stellar mass, SFR and stellarage of green galaxies in various environments of the cosmicweb do not differ in a statistically significant way.Based on the above findings, we conclude that the greenvalley galaxies follow a mixture of evolutionary pathways.We do not find any strong evidence in favour of environmen-tal influences in quenching of star formation in green galax-ies. Only ∼
8% green galaxies in each environment show sig-natures of interactions and merger. 5% of the green galaxiesin each environment are ellipticals which may be the out-come of gas-rich mergers. We also note that the distribu-tions of stellar mass, SFR and stellar age of green galaxiesare independent of their environments.Majority of the green galaxies ( ∼ ∼
10% green galaxies at each environ-ment host AGNs indicates that they play a minor role insuppressing star formation in these galaxies. Dynamical in-stabilities within a galaxy leading to the formation of barand bulge can also suppress star formation within a galaxy.The presence of dominant bulge and bar in ∼
10% greengalaxies at each environment suggests that such internal in- fluences may also play a role of quenching agents in thesegalaxies. However, ∼
70% of the green galaxies must cur-tail their star formation using physical mechanism(s) otherthan interactions, mergers and those driven by bulge, barand AGN activity. We speculate that these green galaxiesare massive galaxies which have grown via secular evolution(Kormendy & Kennicutt 2004) and terminated their starformation via mass driven quenching. However, a long termshutdown of star formation in these galaxies requires thepresence of additional heating sources like AGN feedback,without which accretion and cooling of gas can not be sup-pressed indefinitely. We observe that only a small fraction ofgreen galaxies show AGN activity and hence the quenchingin green galaxies is still a significant problem which requiresother internal physical processes to truncate the gas supplyand maintain the hot halo.Our results are consistent with the findings reported byMendez et al. (2011). They analyzed the AEGIS data overthe redshift range 0 . < z < . The authors thank the SDSS and Galaxy Zoo team for mak-ing the data publicly available. We also greatly acknowledgethe efforts of the Galaxy Zoo and Galaxy Zoo 2 volunteersfor the detailed visual morphological classifications of theSDSS galaxies.BP would like to acknowledge financial support fromthe SERB, DST, Government of India through the projectCRG/2019/001110. BP would also like to acknowledge IU-CAA, Pune for providing support through associateship pro-gramme.Funding for the SDSS and SDSS-II has been providedby the Alfred P. Sloan Foundation, the Participating In-stitutions, the National Science Foundation, the U.S. De-
MNRAS000
MNRAS000 , 1–19 (2020) Das, A., Pandey, B. and Sarkar, S.
REFERENCES
Ahumada R., Allende Prieto C., Almeida A., Anders F., AndersonS. F., Andrews B. H., Anguiano B., et al., 2020, ApJS, 249, 3Alpaslan M., et al., 2014, MNRAS, 438, 177Athanassoula E., Machado R. E. G., Rodionov S. A., 2013, MN-RAS, 429, 1949Baldry I. K., Glazebrook K., Brinkmann J., Ivezić Ž., LuptonR. H., Nichol R. C., Szalay A. S., 2004, ApJ, 600, 681Balogh M. L., Navarro J. F., Morris S. L., 2000, ApJ, 540, 113Balogh M. L., Baldry I. K., Nichol R., Miller C., Bower R., Glaze-brook K., 2004, ApJL, 615, L101Barnes J. E., 2002, MNRAS, 333, 481Bell E. F., McIntosh D. H., Barden M., Wolf C., Caldwell J. A. R.,Rix H.-W., Beckwith S. V. W., et al., 2004, ApJL, 600, L11Bell E. F., Wolf C., Meisenheimer K., Rix H.-W., Borch A., DyeS., Kleinheinrich M., et al., 2004, ApJ, 608, 752Bhattacharjee S., Pandey B., Sarkar S., 2020, JCAP, 2020, 039Binney J., 2004, MNRAS, 347, 1093Birnboim Y., Dekel A., 2003, MNRAS, 345, 349Birnboim Y., Dekel A., Neistein E., 2007, MNRAS, 380, 339Blanton M. R., Brinkmann J., Csabai I., Doi M., Eisenstein D.,Fukugita M., Gunn J. E., et al., 2003, AJ, 125, 2348Bond J. R., Kofman L., Pogosyan D. 1996, Nature, 380, 603Bower R. G., Benson A. J., Malbon R., Helly J. C., Frenk C. S.,Baugh C. M., Cole S., et al., 2006, MNRAS, 370, 645Brammer G. B., Whitaker K. E., van Dokkum P. G., MarchesiniD., Labbé I., Franx M., Kriek M., et al., 2009, ApJL, 706,L173Brinchmann J., Charlot S., White S. D. M., Tremonti C., Kauff-mann G., Heckman T., Brinkmann J., 2004, MNRAS, 351,1151Bruce V. A., Dunlop J. S., Mortlock A., Kocevski D. D., McGrathE. J., Rosario D. J., 2016, MNRAS, 458, 2391 Bruzual A. G., 1983, ApJ, 273, 105.Cameron E., Driver S. P., Graham A. W., Liske J., 2009, ApJ,699, 105Cattaneo A., Dekel A., Devriendt J., Guiderdoni B., Blaizot J.,2006, MNRAS, 370, 1651Cattaneo A., et al., 2007, MNRAS, 377, 63Cimatti A., Brusa M., Talia M., Mignoli M., Rodighiero G., KurkJ., Cassata P., et al., 2013, ApJL, 779, L13Coenda V., Martínez H. J., Muriel H., 2018, MNRAS, 473, 5617Colless, M. et al.(for 2dFGRS team) , 2001,MNRAS, 328, 1039Combes F., Sanders R. H., 1981, A&A, 96, 164Combes F., Gerin M., 1985, A&A, 150, 327Conroy C., Gunn J. E., White M., 2009, ApJ, 699, 486.Correa C. A., Schaye J., Trayford J. W., 2019, MNRAS, 484, 4401Cortese L., 2012, A&A, 543, A132Cox T. J., Primack J., Jonsson P., Somerville R. S., 2004, ApJL,607, L87Croton D. J., Springel V., White S. D. M., De Lucia G., FrenkC. S., Gao L., Jenkins A., et al., 2006, MNRAS, 365, 11Croton D. J., Gao L., White S. D. M., 2007, MNRAS, 374, 1303Darvish, B., Sobral, D., Mobasher, B., et al. 2014, ApJ, 796, 51Debattista V. P., Carollo C. M., Mayer L., Moore B., 2004, ApJL,604, L93Dekel A., Birnboim Y., 2006, MNRAS, 368, 2Dekel A., Birnboim Y., 2008, MNRAS, 383, 119Dekel A., Sari R., Ceverino D., 2009, ApJ, 703, 785Dressler, A., 1980, ApJ, 236, 351Driver S. P., et al., 2006, MNRAS, 368, 414Drory N., et al., 2009, ApJ, 707, 1595Faber S. M., Willmer C. N. A., Wolf C., Koo D. C., Weiner B. J.,Newman J. A., Im M., et al., 2007, ApJ, 665, 265Fabian A. C., 2012, ARA&A, 50, 455Fall S. M., Efstathiou G., 1980, MNRAS, 193, 189Fang J. J., Faber S. M., Koo D. C., Dekel A., 2013, ApJ, 776, 63Filho, M. E., Sánchez Almeida, J., Muñoz-Tuñón, C., et al. 2015,ApJ, 802, 82Fisher D. B., Drory N., 2008, AJ, 136, 773Förster Schreiber N. M., Genzel R., Newman S. F., Kurk J. D.,Lutz D., Tacconi L. J., Wuyts S., et al., 2014, ApJ, 787, 38Gabor J. M., Davé R., Finlator K., Oppenheimer B. D., 2010,MNRAS, 407, 749Gabor J. M., Davé R., 2015, MNRAS, 447, 374Gao L., White S. D. M., 2007, MNRAS, 377, L5Genzel R., Förster Schreiber N. M., Rosario D., Lang P., Lutz D.,Wisnioski E., Wuyts E., et al., 2014, ApJ, 796, 7George K., Subramanian S., Paul K. T., 2019, A&A, 628, A24Gómez P. L., Nichol R. C., Miller C. J., Balogh M. L., Goto T.,Zabludoff A. I., Romer A. K., et al., 2003, ApJ, 584, 210Gunn J. E., Gott J. R., 1972, ApJ, 176, 1Gunn J. E., Carr M., Rockosi C., Sekiguchi M., Berry K., ElmsB., de Haas E., et al., 1998, AJ, 116, 3040Gunn J. E., Siegmund W. A., Mannery E. J., Owen R. E., HullC. L., Leger R. F., Carey L. N., et al., 2006, AJ, 131, 2332Hahn, O., Porciani, C., Carollo, C. M., & Dekel, A. 2007, MN-RAS, 375, 489Hasinger G., 2008, A&A, 490, 905Haywood M., Lehnert M. D., Di Matteo P., Snaith O., SchultheisM., Katz D., Gómez A., 2016, A&A, 589, A66Hogg D. W., Blanton M. R., Eisenstein D. J., Gunn J. E., SchlegelD. J., Zehavi I., Bahcall N. A., et al., 2003, ApJL, 585, L5Hoyle, F., et al. 2002, ApJ, 580, 663Hubble, E.P., 1936, The Realm of the Nebulae (Oxford UniversityPress: Oxford), 79James P. A., Percival S. M., 2018, MNRAS, 474, 3101Jian H.-Y., Lin L., Koyama Y., Tanaka I., Umetsu K., Hsieh B.-C., Higuchi Y., et al., 2020, ApJ, 894, 125Kannappan S. J., 2004, ApJL, 611, L89Kauffmann G., Heckman T. M., White S. D. M., Charlot S.,MNRAS , 1–19 (2020) reen valley in the cosmic web Tremonti C., Peng E. W., Seibert M., et al., 2003, MNRAS,341, 54Kauffmann G., Heckman T. M., White S. D. M., Charlot S.,Tremonti C., Brinchmann J., Bruzual G., et al., 2003, MN-RAS, 341, 33Kauffmann, G., White, S. D. M., Heckman, T. M., et al. 2004,MNRAS, 353, 713Kaviraj S., Peirani S., Khochfar S., Silk J., Kay S., 2009, MNRAS,394, 1713Kawata D., Mulchaey J. S., 2008, ApJL, 672, L103Kereš D., Katz N., Weinberg D. H., Davé R., 2005, MNRAS, 363,2Kerscher M., 2018, A&A, 615, A109Kormendy J., Kennicutt R. C., 2004, ARA&A, 42, 603Kriek M., van der Wel A., van Dokkum P. G., Franx M., Illing-worth G. D., 2008, ApJ, 682, 896Lang P., Wuyts S., Somerville R. S., Förster Schreiber N. M.,Genzel R., Bell E. F., Brammer G., et al., 2014, ApJ, 788, 11Larson R. B., Tinsley B. M., Caldwell C. N., 1980, ApJ, 237, 692Leslie S. K., Kewley L. J., Sanders D. B., Lee N., 2016, MNRAS,455, L82Lewis I., Balogh M., De Propris R., Couch W., Bower R., OfferA., Bland-Hawthorn J., et al., 2002, MNRAS, 334, 673Lin Y.-T., Hsieh B.-C., Lin S.-C., Oguri M., Chen K.-F., TanakaM., Chiu I.-N., et al., 2017, ApJ, 851, 139Lintott, C. J., Schawinski, K., Slosar, A., et al. 2008, MNRAS,389, 1179Luparello, H. E., Lares, M., Paz, D., et al. 2015, MNRAS, 448,1483Madau P., Ferguson H. C., Dickinson M. E., Giavalisco M., SteidelC. C., Fruchter A., 1996, MNRAS, 283, 1388Martig M., Bournaud F., Teyssier R., Dekel A., 2009, ApJ, 707,250Moore B., Katz N., Lake G., Dressler A., Oemler A., 1996, Nature,379, 613Moore B., Lake G., Katz N., 1998, ApJ, 495, 139Masters K. L., Mosleh M., Romer A. K., Nichol R. C., BamfordS. P., Schawinski K., Lintott C. J., et al., 2010, MNRAS, 405,783Menci N., Fontana A., Giallongo E., Salimbeni S., 2005, ApJ, 632,49Mendez A. J., Coil A. L., Lotz J., Salim S., Moustakas J., SimardL., 2011, ApJ, 736, 110Mihos J. C., Hernquist L., 1994, ApJL, 431, L9Miyatake H., More S., Takada M., Spergel D. N., MandelbaumR., Rykoff E. S., Rozo E., 2016, PhRvL, 116, 041301Montero-Dorta A. D., Pérez E., Prada F., Rodríguez-Torres S.,Favole G., Klypin A., Cid Fernandes R., et al., 2017, ApJL,848, L2Murray N., Quataert E., Thompson T. A., 2005, ApJ, 618, 569Musso M., Cadiou C., Pichon C., Codis S., Kraljic K., Dubois Y.,2018, MNRAS, 476, 4877Nandra K., Georgakakis A., Willmer C. N. A., Cooper M. C.,Croton D. J., Davis M., Faber S. M., et al., 2007, ApJL, 660,L11Nelson D., et al., 2018, MNRAS, 475, 624Pandey, B., & Bharadwaj, S. 2006, MNRAS, 372, 827Pandey, B., & Bharadwaj, S. 2008, MNRAS, 387, 767Pandey B., Sarkar S., 2017, MNRAS, 467, L6Pandey B., Sarkar S., 2020, MNRAS, 498, 6069Pandey B., 2020, MNRAS, 499, L31Park, C., et al. 2005, ApJ, 633, 11Planck Collaboration, Aghanim N., Akrami Y., Ashdown M., Au-mont J., Baccigalupi C., Ballardini M., et al., 2020, A&A, 641,A6Postman M., Geller M. J., 1984, ApJ, 281, 95Rees M. J., Ostriker J. P., 1977, MNRAS, 179, 541Salim S., 2014, SerAJ, 189, 1. doi:10.2298/SAJ1489001S Sarkar, P., & Bharadwaj, S. 2009, MNRAS, 394, L66Sarkar S., Pandey B., 2019, MNRAS, 485, 4743Sarkar S., Pandey B., 2020, MNRAS, 497, 4077Schawinski K., Lintott C., Thomas D., Sarzi M., Andreescu D.,Bamford S. P., Kaviraj S., et al., 2009, MNRAS, 396, 818Schawinski K., Urry C. M., Simmons B. D., Fortson L., KavirajS., Keel W. C., Lintott C. J., et al., 2014, MNRAS, 440, 889Scudder, J. M., Ellison, S. L., & Mendel, J. T. 2012, MNRAS,423, 2690Silk J., 1977, ApJ, 211, 638Silverman J. D., Mainieri V., Lehmer B. D., Alexander D. M.,Bauer F. E., Bergeron J., Brandt W. N., et al., 2008, ApJ,675, 1025Somerville R. S., Primack J. R., 1999, MNRAS, 310, 1087Somerville R. S., Hopkins P. F., Cox T. J., Robertson B. E.,Hernquist L., 2008, MNRAS, 391, 481Spinoso D., Bonoli S., Dotti M., Mayer L., Madau P., BellovaryJ., 2017, MNRAS, 465, 3729Springel V., Di Matteo T., Hernquist L., 2005, MNRAS, 361, 776Strauss M. A., Weinberg D. H., Lupton R. H., Narayanan V. K.,Annis J., Bernardi M., Blanton M., et al., 2002, AJ, 124, 1810Strateva I., Ivezić Ž., Knapp G. R., Narayanan V. K., StraussM. A., Gunn J. E., Lupton R. H., et al., 2001, AJ, 122, 1861Taylor E. N., et al., 2015, MNRAS, 446, 2144Thomas D., Maraston C., Schawinski K., Sarzi M., Silk J., 2010,MNRAS, 404, 1775Toomre A., Toomre J., 1972, ApJ, 178, 623Trayford, J. W. et al., 2016, MNRAS, 460, 3925Vakili M., Hahn C., 2019, ApJ, 872, 115Weiner B. J., Phillips A. C., Faber S. M., Willmer C. N. A., VogtN. P., Simard L., Gebhardt K., et al., 2005, ApJ, 620, 595White, S. D. M., & Rees, M. J. 1978, MNRAS, 183, 341Willett K. W., Lintott C. J., Bamford S. P., Masters K. L., Sim-mons B. D., Casteels K. R. V., Edmondson E. M., et al., 2013,MNRAS, 435, 2835Wyder T. K., Martin D. C., Schiminovich D., Seibert M., Bu-davári T., Treyer M. A., Barlow T. A., et al., 2007, ApJS,173, 293York, D. G., et al. 2000, AJ, 120, 1579Zadeh, L. A., 1965, Fuzzy sets. Information and Control, 8, 338This paper has been typeset from a TEX/L A TEX file prepared bythe author.MNRAS000