Nuclear Star Clusters and Bulges
aa r X i v : . [ a s t r o - ph . GA ] A p r Nuclear Star Clusters and Bulges
David R. Cole, Victor P. Debattista
Abstract
Nuclear star clusters are among the densest stellar systems known andare common in both early- and late-type galaxies. They exhibit scaling relationswith their host galaxy which may be related to those of supermassive black holes.These may therefore help us to unravel the complex physical processes occurringat the centres of galaxies. The properties of nuclear stellar systems suggest thattheir formation requires both dissipational and dissipationless processes. They havestellar populations of different ages, from stars as old as their host galaxy to youngstars formed in the last 100 Myr. Therefore star formation must be happening eitherdirectly in the nuclear star cluster or in its vicinity. The secular processes that fuelthe formation of pseudobulges very likely also contributes to nuclear star clustergrowth.
Observations with the high resolution instruments on the
Hubble Space Telescope ( HST ) have revealed that many low to intermediate mass galaxies contain a densestellar system at their centre. They are among the densest stellar systems known.Nuclear stellar systems come in two main morphological types, nuclear star clusters(NSCs), where the stellar distribution is spheroidal, and nuclear discs (NDs). Theseare not mutually exclusive and NSCs often contain a disc component too, typicallycomprised of younger stars. These dense stellar systems are common in galaxiesacross the Hubble sequence.A connection between NSCs and the formation of their host galaxy is impliedby various observed scaling relations between their mass and the properties of their
David R. Cole, Jeremiah Horrocks Institute, University of Central Lancashire, Preston PR1 2HE,e-mail: [email protected] , & Victor P. Debattista, Jeremiah Horrocks Institute, Universityof Central Lancashire, Preston PR1 2HE, e-mail: [email protected] host. These scaling relations provide insight into the physical processes regulatingthe growth of nuclear stellar systems.NSCs in late-type disc galaxies are observed to have a mix of populations, includ-ing young stars formed within the last 100 Myr. Whether these formed in situ, orarrived as star clusters accreted from the neighbourhood of the NSC, gas is neededto make these young stars. Thus NSCs and NDs provide evidence that gas is able torepeatedly reach the centres of late-type galaxies.
NSCs are compact objects with effective radii of order 5 pc and masses rangingfrom 10 M ⊙ to 10 M ⊙ , meaning they have among the highest known average sur-face densities (Walcher et al., 2005). NSCs have photometric and kinematic prop-erties broadly similar to those of globular clusters but with higher velocity disper-sions. Their absolute visual magnitudes lie between -14 and -10 (B¨oker et al., 2002;Cˆot´e et al., 2006) compared with Milky Way globular clusters which have absolutemagnitudes typically in the range -9 to -4 (Harris, 1997). Hartmann et al. (2011)found that the NSC in M33 was photometrically and kinematically consistent withbeing perfectly axisymmetric.NSCs are present in between 50% and 75% of low to intermediate luminositygalaxies. Carollo et al. (1997) found NSCs in 18 of 35 HST
WFPC2 F606W imagesof spiral galaxies including early-types, while B¨oker et al. (2002) found NSCs in 59of 77
HST images of late-type spiral galaxies. Between 66% and 82% of early-typegalaxies in the
HST
ACS Virgo Cluster Survey have NSCs (Cˆot´e et al., 2006), anda similar fraction in the ACS Fornax Cluster Survey (Turner et al., 2012). The pres-ence of a bar does not seem to affect whether a NSCs occurs or not (Carollo et al.,2002; B¨oker et al., 2004).Nuclear discs are also often found in the central regions of galaxies. Theyspan a range of sizes from a few parsecs to of order a kiloparsec in diameter.They can be differentiated from the main galactic disc (if it exists) in that theylie outside of the region where light from the main disc dominates. They arewidely observed in galaxies spanning the full range of Hubble types both in late-type galaxies (Zasov & Moiseev, 1999; Pizzella et al., 2002; Dumas et al., 2007;Garc´ıa-Burillo & Combes, 2012) and in early-types (Scorza & van den Bosch, 1998;Kormendy & Gebhardt, 2001; de Zeeuw et al., 2002; Emsellem et al., 2004; Trujillo et al.,2004; Krajnovi´c et al., 2008; Ledo et al., 2010). Ledo et al. (2010) found that asmany as 20% of early-type galaxies host a nuclear disc. A sample of 48 early-typegalaxies observed as part of the SAURON project, revealed that nuclear discs areassociated with early-type fast rotators (Krajnovi´c et al., 2008). uclear Star Clusters and Bulges 3
NSCs in late-type galaxies often consist of multiple stellar populations. Their meanluminosity-weighted ages range from 10 Myr to 10 Gyr (Rossa et al., 2006), veryoften with evidence of star formation in the last 100 Myr (Walcher et al., 2005,2006). Spectra reveal that their star formation is bursty, with a duty cycle of a fewhundred Myr. For instance, the NSC in M33 had bursts of star formation 40 Myr and1 Gyr ago (Long, Charles & Dubus, 2002). Georgiev & B¨oker (2014), in a study ofNSCs in 228 late-type galaxies, also find their stellar populations span a wide rangeof ages and conclude that recent star formation is ubiquitous.NSCs in late-type disc galaxies are typically elongated approximately in theplane of the main galaxy disc and are often made up of two components, an olderspheroidal component, with a younger and bluer disc embedded in it (Seth et al.,2006, 2008b). In the case of NGC 4244, spectra indicate young ( <
100 Myr) starsin its disc. Integral field spectroscopy reveals that the NSC has rotation in the samesense as the galaxy (see Figure 1) with a relative tilt of 15°. Similarly the NSC ofFCC 277, an elliptical galaxy in the Fornax cluster, is made up of a spheroid and adisc component, both of which are younger than the main galaxy (Lyubenova et al.,2013). Carson et al. (2015) studied
HST
WFC images of 10 of the nearest andbrightest NSCs. They found increasing roundness at longer wavelengths inferringthe existence of blue discs made up of younger stellar populations as in NGC 4244.Most of these NSCs show evidence in colour-colour diagrams of stellar populationsconsisting of a mixture of an older population ( > ∼
80% of its stars more than 5 Gyr ago, with a deep minimum in star formation 1to 2 Gyr ago. The star formation rate then increased again in the last few hundredMyr.Likewise NDs often exhibit a range of ages with a tendency for young starsto be present. Ongoing star formation is observed in the NDs of NGC 5845(Kormendy et al., 1994) and NGC 4486A (Kormendy et al., 2005). In NGC 4486Athe ND manages to have stars more than 2 Gyr younger than the surrounding galaxy(Kormendy et al., 2005). The ND in NGC 4570 shows evidence for recent star for-mation (van den Bosch, Jaffe & van der Marel, 1998). Morelli et al. (2004) foundthat NGC 4478 has a younger stellar population than the main body of the galaxy,with a prolonged star formation history, whereas NGC 4458 has a uniformly oldpopulation. The ND stellar population in NGC 4698 has ages in the range 5 to 10Gyr (Corsini et al., 2012). On the other hand, NDs in early-type galaxies have beenfound to consist mainly of old ( >
10 Gyr) stars (Krajnovi´c & Jaffe, 2004).
One phenomenon which suggests that interactions may play a role in the formationof nuclear stellar systems is kinematic decoupling where distinct stellar components
David R. Cole, Victor P. Debattista have large ( ≥ Fig. 1
The kinematics ofthe NSC in NGC 4244. Top:The measured radial velocityobserved with NIFS. Rota-tion of 30 km s − is clearlyvisible along the major axis.Contours show the K-bandisophotes. The black bar indi-cates 10 pc (0.47 ′′ ). Bottom:Velocity dispersion measure-ments. Figure 2 of Seth et al.(2008b).uclear Star Clusters and Bulges 5 Two principal formation mechanisms have been advanced to explain the forma-tion of NSCs. The first is that NSCs form due to globular clusters falling to thecentres of galaxies under the action of dynamical friction and subsequently merge(Tremaine, Ostriker & Spitzer, 1975; Capuzzo-Dolcetta, 1993; Miocchi et al., 2006;Capuzzo-Dolcetta & Miocchi, 2008b,a; Antonini et al., 2012; Antonini, 2013; Gnedin, Ostriker & Tremaine,2014). A candidate infalling globular cluster was found in the inner few hun-dred parsecs of NGC 2139 which could become a NSC in a few hundred Myr(Andersen et al., 2008). An off-centre super star cluster with a mass of 1 . + . − . × M ⊙ has been observed in NGC 253. This super star cluster is a candidate futureNSC (Kornei & McCrady, 2009). Georgiev & B¨oker (2014) present NGC 4654 asan example of galaxies where two star clusters are present at the centre. The two starclusters have a mass ratio of order 10 : 1 and are separated by ∼
30 pc (in projection).The less massive of the star clusters appears to be young ( <
100 Myr), supportingthe picture of NSC growth due to the accretion of young globular clusters onto thecentre. Nguyen et al. (2014) find a starburst at the centre of a SMBH-hosting galaxyHeinze 2-10. This starburst has created several super star clusters within 100 pcof the SMBH and they conclude that the star clusters would merge due to dynam-ical friction and form an NSC in the next Gyr. Simulations by Capuzzo-Dolcetta(1993) showed that, while the infall of globular clusters is in competition with tidalstripping (which destroys the clusters), nonetheless a fraction of them do manage toreach the nucleus. Fewer massive globular clusters than expected have been foundin the inner region of dwarf ellipticals suggesting that the globular clusters missingfrom the inner regions had been depleted due to their shorter dynamical friction in-fall times and merging to form NSCs (Lotz et al., 2001; Lotz, Miller & Ferguson,2004). The infall of globular clusters has been modelled and simulated severaltimes. Agarwal & Milosavljevi´c (2011) modelled the infall (and stripping) of glob-ular clusters analytically and concluded that this process could create NSCs whichmatch the observed NSC masses. Capuzzo-Dolcetta & Miocchi (2008a) show thatglobular clusters merging at the centre of a bulge leads to density-velocity dispersionproperties consistent with those of observed NSCs. Analytic modelling of the infallof globular clusters led Gnedin, Ostriker & Tremaine (2014) to the same conclu-sion; they argue, moreover, that the contrast between the NSC and the backgroundgalaxy would be much lower in a massive galaxy like M87, making them harder todetect in such systems. den Brok et al. (2014) studied NSCs in 200 Coma clusterdwarf elliptical galaxies and found a relation between NSC and host galaxy mag-nitude of M nuc = ( . ± . )( M gal + . ) − ( . ± . ) concluding that thisis consistent with predictions of how NSC luminosity scales with host galaxy lu-minosity as predicted by the globular cluster merger scenario models of Antonini(2013) and Gnedin, Ostriker & Tremaine (2014). However they also find that galax-ies with higher S´ersic indices tend to have brighter NSCs. They argue that this is David R. Cole, Victor P. Debattista due to the brighter galaxies being better able to retain gas and conclude that in situstar formation also plays an important role.den Brok et al. (2014) showed an example of an unnucleated galaxy with severalold globular clusters within it which raises the question of how this galaxy managedto avoid forming a NSC. Hartmann et al. (2011) note that the NSC in M33 appearsaxisymmetric (both photometrically and kinematically); their simulations of glob-ular cluster mergers produce triaxial NSCs although this outcome can be avoidedif a massive black hole is present. However in M33 the upper limit on the pres-ence of a black hole is very stringent (M bh < M ⊙ , (Merritt & Ferrarese, 2001;Gebhardt et al., 2001)) suggesting that its NSC did not form via globular clustermergers. On the other hand, the NSC in the Milky Way has a rotating sub-structureperpendicular to the Galactic plane (Feldmeier et al., 2014), suggesting a cluster wasaccreted. in situ star formation has been proposed as an alternative for forming NSCs(Milosavljevi´c, 2004; Bekki, 2007). Cen (2001) suggested that at the epoch of re-ionisation the external radiation field could create an inward convergent shock lead-ing to the formation of massive dense clusters at the centres of early galaxies withmasses and velocity dispersions comparable to those of NSCs. While these couldhave formed the seeds of some NSCs, the NSCs would have had to grow furthersince formation to account for the younger populations. Emsellem & van de Ven(2008) showed that the tidal field of a wide range of S´ersic profile spheroids are com-pressive in the regions where NSCs form; gas falling in is therefore likely to formstars. They found that the mass of the object expected to form would be 0 . − . v rms = q s + v , which is not observed in the data. Based on this,they conclude that less than 50% of the mass of the NSC could have been assembledfrom the mergers of globular clusters, with the majority due to in situ star formation.On the other hand, they also find a negative vertical anisotropy, b z = − s z / s R ,confirmed through the independent modelling of De Lorenzi et al. (2013) using themade-to-measure technique (Syer & Tremaine, 1996; de Lorenzi et al., 2007). This,they showed, could be produced by the accretion of a globular cluster, accountingfor at least 10% of its mass, on a nearly polar orbit relative to the NSC. They con-clude that both in situ star formation and globular clusters mergers played a role inthe formation of this NSC. From a sample of over 200 late-type spiral galaxies ob-served with HST , Georgiev & B¨oker (2014) showed that NSCs are smaller in bluecompared to red filters. This can be explained either by the presence of an AGNor by population gradients within the NSC, possibly indicating ongoing star forma-tion. Turner et al. (2012) studied the nuclei in 43 early-type galaxies in the Fornaxcluster. On the basis of globular cluster infall times, they concluded that in low massearly-type galaxies the dominant mechanism for NSC formation is probably glob-ular cluster merging but for more massive galaxies in situ star formation becomesnecessary. A picture is being established therefore where both processes, globular uclear Star Clusters and Bulges 7 cluster mergers and in situ star formation, play a role in NSC formation and which isthe dominant mechanism depends on the parameters of the host galaxy (Rossa et al.,2006; Walcher et al., 2006).
The formation of NDs is thought to require in situ star formation. A significant massof gas needs to be funnelled to the nuclear regions to allow this. Such inflows arepossible in mergers as shown by hydrodynamical simulations (Mayer, Kazantzidis & Escala,2008; Mayer et al., 2010; Hopkins & Quataert, 2010; Chapon, Mayer & Teyssier,2013). Chapon, Mayer & Teyssier (2013) presented a simulation of the merger of
Fig. 2
Real and simulated NSCs on the (V / s , e ) diagram of Binney (2005). The green stars andred triangles indicate the location of the simulated NSCs (M1-M3 & A1-A3) formed by merginginfalling globular clusters, while the magenta squares are the observed NSCs in NGC 4244 andM33. The observed location for M33 has been projected (red line) assuming an inclination of i = two galaxies with SMBHs (see Figure 3). After the merger a thick nuclear gas discforms with a mass ∼ M ⊙ . Nuclear discs have been observed in 17 nearby lumi-nous infra-red galaxies (LIRGs) and ultra-luminous infra-red galaxies (ULIRGs),possibly the results of merger-driven gas funnelling to their centres initiating in-tense star formation (Medling et al., 2014). Meanwhile Hopkins & Quataert (2010)showed that lopsided nuclear discs such as the one in the Andromeda galaxy mayform via merger-driven inflows. Instead Ledo et al. (2010) showed that pre-existingNDs are destroyed in mergers.Secular processes such as the action of a bar can also supply gas to form aND. The formation of the ND in the edge-on galaxy NGC 7332 was attributedto the presence of a bar (which was inferred from the boxy/peanut-shaped bulge)(Seifert & Scorza, 1996; Falc´on-Barroso et al., 2004). Barred galaxies have moremolecular gas in their central kiloparsec than unbarred galaxies (Sakamoto et al.,1999; Sheth et al., 2005). Enhanced nuclear star formation correlates with the pres-ence of a strong bar in disc galaxies (Wang et al., 2012), and depends primarily onthe ellipticity of the bar, not on the size of the bar. However only half of galaxieswith centrally concentrated star formation have a strong bar suggesting that pro- Fig. 3
Gas density mapsduring the final stages of themerger of two galaxies. Weare looking down onto theorbital plane of the galaxiesand the maps are 1.8 kpc wide.Gas from the two galaxiesfunnels inwards to form athick gaseous nuclear discwith two SMBHs orbiting init in the final image. Figure 1of Chapon, Mayer & Teyssier(2013)uclear Star Clusters and Bulges 9 cesses such as interactions with other galaxies also induce star formation in thenucleus.Cole et al. (2014) presented a simulation of the formation of an L ∗ isolatedgalaxy. After the bar formed, a ND developed (see Figure 4). They demonstratedthat gas flows to the centre and fuels star formation. The resulting ND is elongatedperpendicular to the main bar, suggesting that the stars in the ND are on x2 orbits.The ND can clearly be seen in the kinematics and the stellar metallicity.Given the available data, ND formation through dissipationless processes cannotbe excluded. Agarwal & Milosavljevi´c (2011) proposed that NDs form out of thedebris of infalling star clusters and Portaluri et al. (2013) showed that such a sce-nario is consistent with the available kinematic and photometric data. It has alsobeen demonstrated that NDs can be formed from accreted dwarf satellites settlinginto rotationally supported NDs (Eliche-Moral et al., 2011). Just like NSCs and NDs, pseudobulges require the inflow of gas to form. Pseudob-ulges have proven to be very common, with Fisher & Drory (2011) estimating that
Fig. 4
Face-on (bottom) and edge-on (top) stellar surface density for young stars ( < L ∗ galaxy. The galaxy has a bar which is oriented along the x-axis.A thin disc of stars can clearly be seen. Figure 4 of Cole et al. (2014)0 David R. Cole, Victor P. Debattista they account for ∼
80% of disc galaxies. Kormendy & Kennicutt (2004) reviewedthe formation of pseudobulges via the funnelling of gas through non-axisymmetricstructures, such as bars. They list eight key properties of NSCs which need to beunderstood if NSCs and pseudobulges are related secular phenomena: (1) NSCs arecommon (2) NSCs are rare in irregulars (3) NSCs are fairly homogeneous in theirproperties (4) NSCs are at the centres of their host galaxies (5) NSCs host youngstars (6) NSCs are not more common in barred galaxies (7) In the FundamentalPlane NSCs are more similar to globular clusters and (8) The masses of NSCs cor-relate with the luminosities of their host galaxies. They argue that points (2), (3), (6)and (7) appear inconsistent with NSCs and pseudobulges being related phenomena.Possible answers to these problems could be that the centres of irregulars are notwell-defined (point 2), that globular cluster mergers are responsible for part of themass assembly of NSCs (points 3 and 7) and the gas flows required for NSCs arenot as large as needed for pseudobulges (point 6). Nonetheless, NSCs are different.Walcher et al. (2005) tested the idea that NSCs are proto-bulges which form throughin situ star formation but whose growth has not been sufficient to form a bulge. Onthe basis of 9 bulgeless galaxies, they showed that the dynamical properties of theNSCs are very different to those bulges.How can gas get funnelled to NSCs? Kormendy & Kennicutt (2004) invoked barsand ovals to explain the formation of pseudobulges. It has long been recognised that,while bars can drive gas inwards, this gas stalls at the inner Lindblad resonance.Shlosman, Frank & Begelman (1989) proposed that gas can be driven all the wayto the centre of a galaxy, thereby feeding AGN activity, by means of nested bars,where a small-scale bar resides inside a larger bar. Such double-barred galaxies havebeen observed in about 25% of early-type galaxies (Erwin & Sparke, 2002). In thisscenario the main bar of a galaxy would induce an inward flow creating a nucleargas disc which could again become unstable leading to further gas infall. Evidencefor gas inflow that can be explained by this scenario comes from the molecular gasin NGC 6946 (Schinnerer et al., 2006, 2007), which appears to be streaming alongthe leading edge of an inner stellar bar about 400 pc long nested inside a large-scale(3.5 kpc) bar.However, the fact that NSCs do not prefer barred galaxies ((Carollo et al., 2002;B¨oker et al., 2004)) suggests that bars are not the sole mechanism responsible forfunnelling gas to nuclei. As an alternative, Milosavljevi´c (2004) proposed that themagneto-rotational instability could transport neutral gas inside 100 pc where itcould form stars.NDs can sometimes be directly associated with pseudobulges through the phe-nomenon of s -drops, where galaxies have a significant drop in velocity dispersionin their centre (e.g. Emsellem et al., 2001). These can be explained by infalling gasforming a dynamically cool ND. Star formation reduces the central velocity dis-persion (Wozniak et al., 2003; Comer´on, Knapen & Beckman, 2008). Small NDshave been observed with HST in the centre of galaxies co-located with s -drops(Mendez-Abreu et al., 2014). uclear Star Clusters and Bulges 11 A small fraction of galaxies host both a NSC and a SMBH (Seth et al., 2008a), al-though the actual fraction could be higher given the difficulties in detecting both ina given galaxy. Neumayer & Walcher (2012) noted that a plot of M bh versus M NSC divides into three regions, one which is NSC dominated, a transition region and onewhich is SMBH dominated. This led them to speculate that SMBHs form insideNSCs but outgrow and destroy them when the NSC mass is less than one percentof the SMBH mass. Alternatively, Nayakshin, Wilkinson & King (2009) proposed acompetitive feedback to explain the dichotomy between NSCs and SMBHs. Argu-ing that NSC growth depends on the dynamical time of the nuclear region, they findthat there is a transition when the velocity dispersion of the host spheroid is ∼ − . Above this the NSC cannot grow efficiently and below this the SMBHcannot grow efficiently thus explaining why NSCs are mainly found in low and in-termediate mass galaxies. Antonini et al. (2012), and Antonini (2013) explained thisdichotomy in the globular cluster formation scenario in the presence of a SMBH.For a low mass SMBH, such as the Milky Way’s, globular clusters manage to reachthe centre allowing the NSC to grow. However, if M bh ∼ M ⊙ then the globularclusters are disrupted before reaching the nucleus.However Kormendy & Ho (2013) point out that there is no segregation into gi-ants that only contain SMBHs or dwarfs that only contain nuclei. Where SMBHsand NSCs co-exist the ratio of SMBH to NSC mass can vary across a large, and apparently continuous range above and below unity. For example in NGC 4026 M bh M NSC = . M bh M NSC = . ± . The masses of SMBHs, M bh are well known to correlate with their host galaxyproperties including the bulge velocity dispersion, s e (Ferrarese & Merritt, 2000;Gebhardt et al., 2000; Merritt & Ferrarese, 2001; Tremaine et al., 2002; Ferrarese & Ford,2005; G¨ultekin et al., 2009; Graham et al., 2011; McConnell et al., 2011; Beifiori et al.,2012), the bulge mass, M bul (Magorrian et al., 1998; Marconi & Hunt, 2003; H¨aring & Rix,2004; Sani et al., 2011; Beifiori et al., 2012; Graham, 2012) and the bulge luminos-ity, L bul (Kormendy & Richstone, 1995; McLure & Dunlop, 2002; Marconi & Hunt,2003; Graham, 2007; G¨ultekin et al., 2009; Sani et al., 2011; McConnell et al., 2011;Beifiori et al., 2012; Graham & Scott, 2013). (See Chapter 5 of this volume “Galaxy bulges and their massive black holes” by Alister W. Graham for an up-to-date dis-cussion on SMBH scaling relations and the review by Kormendy & Ho (2013).)Similarly the luminosity and mass of nuclear star clusters and nuclear discs havebeen found to correlate with their host galaxy properties (Balcells et al., 2003;Balcells, Graham & Peletier, 2007; Graham & Guzm´an, 2003; Ferrarese et al., 2006;Wehner & Harris, 2006; Graham, 2012) which has led to them being lumped alongwith SMBHs as a generic class of objects, the central massive objects (CMOs), withstellar CMOs being found in less massive galaxies.The luminosity of stellar CMOs was found to correlate with that of their hostbulge in disc galaxies by (Balcells et al., 2003; Balcells, Graham & Peletier, 2007)and a similar correlation was found in a study of dE galaxies in the Coma clusterby (Graham & Guzm´an, 2003). Ferrarese et al. (2006) found that the mass of stellarCMOs in early-type galaxies correlates with both the host galaxy’s luminosity anddynamical mass, M galdyn (cid:181) s e R gale (Figure 5). More importantly they found a commonM CMO -M gal relationship for galaxies with either NSCs or SMBHs, with NSCs occu-pying fainter galaxies with lower s e . This common relation suggests that there is asingle mechanism responsible for regulating the growth of CMOs. They speculatedthat stellar nuclei form in all galaxies but in the most massive ones they collapseto a SMBH. A similar common relation between CMO mass and the mass of thehost galaxy was also found by Wehner & Harris (2006) in dwarf elliptical galaxies.McLaughlin, King & Nayakshin (2006) proposed that the mechanism responsiblefor this correlation is momentum-driven feedback, from supernovae in the case ofNSCs and from AGN activity in the case of SMBHs.However the existence of scaling relations in common between NSCs andSMBHs has recently been questioned. The first indication that stellar CMOs andSMBHs do not, in fact, follow the same scaling relations came in a study ofS0-Sbc galaxies by Balcells, Graham & Peletier (2007). They found that the nearinfra-red luminosities of NSCs scale with host bulge luminosities. However, incontrast to Ferrarese et al. (2006), when they added SMBHs they found a nonlin-ear dependence between M CMO and M bulge . An expanded dataset allowed Graham(2012) to show that NSC mass correlates with host spheroid velocity dispersion as log [ M NSC / M ⊙ ] = . ± .
24 log [ s / kms − ] + ( . ± . ) . The slope of thisrelation, ∼
2, is much lower than that for SMBHs, ∼
5. Leigh, B¨oker & Knigge(2012) found that NSC mass is directly proportional to host spheroid mass; thevirial theorem then implies an M
NSC - s e relation with a slope again close to 2. Fig-ure 6, taken from Scott & Graham (2013), shows their relations between CMO massversus galaxy magnitude, s and galaxy virial mass; these relations can be com-pared directly with those of Ferrarese et al. (2006), shown in Figure 5 find a slopeof 2 . ± . M NSC − s relation, much shallower than the relation found byFerrarese et al. (2006). A major reason for this difference is their inclusion of NSCsin more massive galaxies and the exclusion of NDs. Erwin & Gadotti (2012) andScott & Graham (2013) both noted that the mass of NSCs correlates better withthe host’s total stellar mass, whereas that of SMBHs correlates better with the hostspheroid. They conclude that different physical processes regulate NSC and SMBHgrowth. uclear Star Clusters and Bulges 13 Kormendy & Ho (2013) reach a more nuanced conclusion on the relation be-tween SMBHs and NSCs by taking galaxy type into account when studying theratio of CMO to bulge or galaxy mass. NSCs in spheroidal galaxies are rela-tively more massive than in late-type galaxies, consistent with the generally heldview that spheroidal galaxies are late-type galaxies which have lost baryonic mass.This renders these galaxies less useful for comparing CMO scaling relations. Theyfind that the ratio ( M bh + M NSC ) / M bulge has less scatter than either M bh / M bulge or M NSC / M bulge , suggesting that the evolution of NSCs and SMBHs is tightly coupled.As a fraction of total galaxy mass, instead, both SMBHs and NSCs have a largerrelative mass in early-type galaxies compared with late-type galaxies (excluding thespheroidals). Kormendy & Ho (2013) conclude that this hints at SMBHs and NSCsbeing related.Although a common scaling relation between NSCs and SMBHs now seemsdead, nonetheless the existence of scaling relations between NSCs and their hostgalaxies still provide important constraints on how their growth is regulated (Silk & Rees,1998; King, 2003; Wyithe & Loeb, 2003; Di Matteo, Springel & Hernquist, 2005;Murray, Quataert & Thompson, 2005; Sazonov et al., 2005; Younger et al., 2008;Booth & Schaye, 2009; Johansson, Naab & Burkert, 2009; Power et al., 2011; Wyithe & Loeb,2003; Di Matteo, Springel & Hernquist, 2005; Murray, Quataert & Thompson, 2005;Sazonov et al., 2005; Younger et al., 2008; Johansson, Naab & Burkert, 2009). The-oretical models of formation mechanisms make predictions for scaling relations be-tween NSCs and host galaxy properties, based on underlying physics, and these al- Fig. 5
The mass of the CMO versus the blue band magnitude (left), velocity dispersion s (mid-dle) and host spheroid dynamical mass (right) for SMBHs (filled and open circles) and nuclei(red squares). The solid red and black lines show the best fits to the nuclei and early-type SMBHsamples, respectively. In the right panel, the dashed line is the fit obtained for the combined nu-clei+SMBH sample. Figure 2 of Ferrarese et al. (2006)4 David R. Cole, Victor P. Debattista low us to distinguish how NSCs are formed. The main process proposed is feedbackfrom the CMO, and possibly the effect this has on its galaxy but the exact mecha-nisms is not yet clear (Silk & Rees, 1998; Springel, Di Matteo & Hernquist, 2005;Booth & Schaye, 2009; Fabian, 1999; King, 2003, 2005; Murray, Quataert & Thompson,2005; McLaughlin, King & Nayakshin, 2006; King, 2010; Power et al., 2011; McQuillin & McLaughlin,2012). Observations of nuclear star clusters show that these systems can unravel the massassembly at the centres of galaxies. Their properties give us clues as to how theywere formed and the physical processes that contributed to their formation. There isa close connection between the the formation of NSCs and their host, as is demon-strated by their scaling relations. Observations support both in situ star formationand globular cluster merging for the formation and growth of NSCs. How muchthese mechanisms contribute to the growth of NSCs probably depends on whetherthey are found in early or late-type galaxies or in high-mass or low-mass galaxies.It also seems likely that the morphology of the bulge, whether it is a classical bulge
Fig. 6
The mass of the CMO versus the B band magnitude, velocity dispersion s and galaxydynamical mass for SMBHs (black dots), NDs (open symbols) and NSCs (red dots). The solidblack lines shows the slope of the relations for SMBHs and the solid red lines for NSCs. Figure 2 ofScott & Graham (2013). The scaling relation for SMBHs is updated in Figure 2 of Graham & Scott(2013), and Figure 3 of Scott, Graham & Schombert (2013).uclear Star Clusters and Bulges 15 or a discy bulge formed through secular processes, will affect the transport of gas tothe nuclear regions of a galaxy where it can form stars. NSCs in early-type galaxieswith little gas are unlikely to grow due to dissipational processes. However late-typegalaxies show multiple stellar populations with stars of the order of a few hundredMyr old implying recent star formation.NSCs and SMBHs can co-exist, as can be seen in the Milky Way, but they nolonger seem to be two types of a single central massive object. However studyingthe interrelationship between these two types of nuclear system will contribute tothe understanding of all physical processes which are important in their formation.Further investigation of how scaling relations are affected by the presence of a bar,the morphology of the bulge and whether the host galaxy is early- or late-type mayallow us to refine our ideas of how they form. There is also a need for higher resolu-tion simulations of the effects of in situ star formation on the kinematics, chemistryand morphology of NSCs. From the information we have gathered so far it is clearthat NSC formation is very complicated and their continued study will bring furtherinsights. Acknowledgements
DRC and VPD are supported by STFC Consolidated grant ST/J001341/1.
References
Agarwal M., Milosavljevi´c M., 2011, ApJ, 729, 35Andersen D. R., Walcher C. J., B¨oker T., Ho L. C., van der Marel R. P., Rix H.-W.,Shields J. C., 2008, ApJ, 688, 990Antonini F., 2013, ApJ, 763, 62Antonini F., Capuzzo-Dolcetta R., Mastrobuono-Battisti A., Merritt D., 2012, ApJ,750, 111Balcells M., Graham A. W., Dom´ınguez-Palmero L., Peletier R. F., 2003, ApJ, 582,L79Balcells M., Graham A. W., Peletier R. F., 2007, ApJ, 665, 1084Beifiori A., Courteau S., Corsini E. M., Zhu Y., 2012, MNRAS, 419, 2497Bekki K., 2007, PASA, 24, 77Bertola F., Corsini E. M., 1999, in IAU Symposium, Vol. 186, Galaxy Interactionsat Low and High Redshift, Barnes J. E., Sanders D. B., eds., p. 149Binney J., 2005, MNRAS, 363, 937B¨oker T., Laine S., van der Marel R. P., Sarzi M., Rix H.-W., Ho L. C., Shields J. C.,2002, AJ, 123, 1389B¨oker T., Sarzi M., McLaughlin D. E., van der Marel R. P., Rix H.-W., Ho L. C.,Shields J. C., 2004, AJ, 127, 105Booth C. M., Schaye J., 2009, MNRAS, 398, 53Cappellari M. et al., 2007, MNRAS, 379, 418Capuzzo-Dolcetta R., 1993, ApJ, 415, 616
Capuzzo-Dolcetta R., Miocchi P., 2008a, ApJ, 681, 1136Capuzzo-Dolcetta R., Miocchi P., 2008b, MNRAS, 388, L69Carollo C. M., Stiavelli M., de Zeeuw P. T., Mack J., 1997, AJ, 114, 2366Carollo C. M., Stiavelli M., Seigar M., de Zeeuw P. T., Dejonghe H., 2002, AJ, 123,159Carson D. J., Barth A. J., Seth A. C., den Brok M., Cappellari M., Greene J. E., HoL. C., Neumayer N., 2015, ArXiv e-printsCen R., 2001, ApJ, 560, 592Chapon D., Mayer L., Teyssier R., 2013, MNRAS, 429, 3114Cole D. R., Debattista V. P., Erwin P., Earp S. W. F., Roˇskar R., 2014, MNRAS, 445,3352Comer´on S., Knapen J. H., Beckman J. E., 2008, A&A, 485, 695Corsini E. M., M´endez-Abreu J., Pastorello N., Dalla Bont`a E., Morelli L., BeifioriA., Pizzella A., Bertola F., 2012, MNRAS, 423, L79Corsini E. M. et al., 1999, A&A, 342, 671Cˆot´e P. et al., 2006, ApJS, 165, 57de Lorenzi F., Debattista V. P., Gerhard O., Sambhus N., 2007, MNRAS, 376, 71De Lorenzi F., Hartmann M., Debattista V. P., Seth A. C., Gerhard O., 2013, MN-RAS, 429, 2974de Zeeuw P. T. et al., 2002, MNRAS, 329, 513den Brok M. et al., 2014, MNRAS, 445, 2385Di Matteo T., Springel V., Hernquist L., 2005, Nature, 433, 604Dumas G., Mundell C. G., Emsellem E., Nagar N. M., 2007, MNRAS, 379, 1249Eliche-Moral M. C., Gonz´alez-Garc´ıa A. C., Balcells M., Aguerri J. A. L., GallegoJ., Zamorano J., Prieto M., 2011, A&A, 533, A104Emsellem E. et al., 2004, MNRAS, 352, 721Emsellem E., Greusard D., Combes F., Friedli D., Leon S., P´econtal E., WozniakH., 2001, A&A, 368, 52Emsellem E., van de Ven G., 2008, ApJ, 674, 653Erwin P., Gadotti D. A., 2012, Advances in Astronomy, 2012Erwin P., Sparke L. S., 2002, AJ, 124, 65Fabian A. C., 1999, MNRAS, 308, L39Falc´on-Barroso J. et al., 2004, MNRAS, 350, 35Feldmeier A. et al., 2014, A&A, 570, A2Ferrarese L. et al., 2006, ApJ, 644, L21Ferrarese L., Ford H., 2005, Space Sci. Rev. , 116, 523Ferrarese L., Merritt D., 2000, ApJ, 539, L9Fisher D. B., Drory N., 2011, ApJ, 733, L47Garc´ıa-Burillo S., Combes F., 2012, Journal of Physics Conference Series, 372,012050Gebhardt K. et al., 2000, ApJ, 539, L13Gebhardt K. et al., 2001, AJ, 122, 2469Georgiev I. Y., B¨oker T., 2014, MNRAS, 441, 3570Gnedin O. Y., Ostriker J. P., Tremaine S., 2014, ApJ, 785, 71Graham A. W., 2007, MNRAS, 379, 711 uclear Star Clusters and Bulges 17
Graham A. W., 2012, MNRAS, 422, 1586Graham A. W., Guzm´an R., 2003, AJ, 125, 2936Graham A. W., Onken C. A., Athanassoula E., Combes F., 2011, MNRAS, 412,2211Graham A. W., Scott N., 2013, ApJ, 764, 151G¨ultekin K. et al., 2009, ApJ, 698, 198H¨aring N., Rix H.-W., 2004, ApJ, 604, L89Harris W. E., 1997, VizieR Online Data Catalog, 7202, 0Hartmann M., Debattista V. P., Seth A., Cappellari M., Quinn T. R., 2011, MNRAS,418, 2697Hopkins P. F., Quataert E., 2010, MNRAS, 407, 1529Johansson P. H., Naab T., Burkert A., 2009, ApJ, 690, 802King A., 2003, ApJ, 596, L27King A., 2005, ApJ, 635, L121King A. R., 2010, MNRAS, 402, 1516Kormendy J., Dressler A., Byun Y. I., Faber S. M., Grillmair C., Lauer T. R., Rich-stone D., Tremaine S., 1994, in European Southern Observatory Conference andWorkshop Proceedings, Vol. 49, European Southern Observatory Conference andWorkshop Proceedings, Meylan G., Prugniel P., eds., p. 147Kormendy J., Gebhardt K., 2001, in American Institute of Physics Conference Se-ries, Vol. 586, 20th Texas Symposium on relativistic astrophysics, Wheeler J. C.,Martel H., eds., pp. 363–381Kormendy J., Gebhardt K., Fisher D. B., Drory N., Macchetto F. D., Sparks W. B.,2005, AJ, 129, 2636Kormendy J., Ho L. C., 2013, ARA&A, 51, 511Kormendy J., Kennicutt, Jr. R. C., 2004, ARA&A, 42, 603Kormendy J., Richstone D., 1995, ARA&A, 33, 581Kornei K. A., McCrady N., 2009, ApJ, 697, 1180Krajnovi´c D. et al., 2008, MNRAS, 390, 93Krajnovi´c D., Jaffe W., 2004, A&A, 428, 877Lauer T. R. et al., 2005, AJ, 129, 2138Launhardt R., Zylka R., Mezger P. G., 2002, A&A, 384, 112Ledo H. R., Sarzi M., Dotti M., Khochfar S., Morelli L., 2010, MNRAS, 407, 969Leigh N., B¨oker T., Knigge C., 2012, MNRAS, 424, 2130Long K. S., Charles P. A., Dubus G., 2002, ApJ, 569, 204Lotz J. M., Miller B. W., Ferguson H. C., 2004, ApJ, 613, 262Lotz J. M., Telford R., Ferguson H. C., Miller B. W., Stiavelli M., Mack J., 2001,ApJ, 552, 572Lyubenova M. et al., 2013, MNRAS, 431, 3364Magorrian J. et al., 1998, AJ, 115, 2285Marconi A., Hunt L. K., 2003, ApJ, 589, L21Mayer L., Kazantzidis S., Escala A., 2008, Mem. Societa Astronomica Italiana, 79,1284Mayer L., Kazantzidis S., Escala A., Callegari S., 2010, Nature, 466, 1082
McConnell N. J., Ma C.-P., Gebhardt K., Wright S. A., Murphy J. D., Lauer T. R.,Graham J. R., Richstone D. O., 2011, Nature, 480, 215McDermid R. M. et al., 2006, MNRAS, 373, 906McLaughlin D. E., King A. R., Nayakshin S., 2006, ApJ, 650, L37McLure R. J., Dunlop J. S., 2002, MNRAS, 331, 795McQuillin R. C., McLaughlin D. E., 2012, MNRAS, 423, 2162Medling A. M. et al., 2014, ApJ, 784, 70Mendez-Abreu J., Debattista V. P., Corsini E. M., Aguerri J. A. L., 2014, ArXive-printsMerritt D., 2009, ApJ, 694, 959Merritt D., Ferrarese L., 2001, ApJ, 547, 140Milosavljevi´c M., 2004, ApJ, 605, L13Miocchi P., Capuzzo Dolcetta R., Di Matteo P., Vicari A., 2006, ApJ, 644, 940Morelli L., Cesetti M., Corsini E. M., Pizzella A., Dalla Bont`a E., Sarzi M., BertolaF., 2010, A&A, 518, A32Morelli L. et al., 2004, MNRAS, 354, 753Murray N., Quataert E., Thompson T. A., 2005, ApJ, 618, 569Nayakshin S., Wilkinson M. I., King A., 2009, MNRAS, 398, L54Neumayer N., Walcher C. J., 2012, Advances in Astronomy, 2012, 15Nguyen D. D., Seth A. C., Reines A. E., den Brok M., Sand D., McLeod B., 2014,ApJ, 794, 34Pfuhl O. et al., 2011, ApJ, 741, 108Pizzella A., Corsini E. M., Morelli L., Sarzi M., Scarlata C., Stiavelli M., Bertola F.,2002, ApJ, 573, 131Portaluri E., Corsini E. M., Morelli L., Hartmann M., Dalla Bont`a E., DebattistaV. P., Pizzella A., 2013, MNRAS, 433, 434Power C., Zubovas K., Nayakshin S., King A. R., 2011, MNRAS, 413, L110Quinlan G. D., 1996, New Astronomy, 1, 255Rossa J., van der Marel R. P., B¨oker T., Gerssen J., Ho L. C., Rix H.-W., ShieldsJ. C., Walcher C.-J., 2006, AJ, 132, 1074Sakamoto K., Okumura S. K., Ishizuki S., Scoville N. Z., 1999, ApJ, 525, 691Sani E., Marconi A., Hunt L. K., Risaliti G., 2011, MNRAS, 413, 1479Sazonov S. Y., Ostriker J. P., Ciotti L., Sunyaev R. A., 2005, MNRAS, 358, 168Schinnerer E., B¨oker T., Emsellem E., Downes D., 2007, A&A, 462, L27Schinnerer E., B¨oker T., Emsellem E., Lisenfeld U., 2006, ApJ, 649, 181Scorza C., van den Bosch F. C., 1998, MNRAS, 300, 469Scott N., Graham A. W., 2013, ApJ, 763, 76Scott N., Graham A. W., Schombert J., 2013, ApJ, 768, 76Seifert W., Scorza C., 1996, A&A, 310, 75Seth A., Ag¨ueros M., Lee D., Basu-Zych A., 2008a, ApJ, 678, 116Seth A. C., Blum R. D., Bastian N., Caldwell N., Debattista V. P., 2008b, ApJ, 687,997Seth A. C., Dalcanton J. J., Hodge P. W., Debattista V. P., 2006, AJ, 132, 2539Sheth K., Vogel S. N., Regan M. W., Thornley M. D., Teuben P. J., 2005, ApJ, 632,217 uclear Star Clusters and Bulges 19uclear Star Clusters and Bulges 19