WiFeS spectroscopy allowed us to study in detail the physical conditions of the gas of the galaxy SOS 114372, which is currently experiencing ram-pressure stripping (see here).

Fig. 1 – Areas in which the spatial bins of WiFeS have been averaged to obtain enough signal-to-noise ratio to achieve reliable flux ratios.

The flux measured ratios allowed us to obtain the diagnostic diagrams shown in Fig. 2. Note that the colours of the points in Fig. 2 correspond to the colours of the areas in Fig. 1, so that the physical condition of each area can be immediately established.

Fig. 2 – Line flux diagnostic diagrams of the different regions of SOS 114372 (see text). The shock and photoionization models by Rich et al. (2011) are superimposed on the measured flux ratios. The black curves are drawn for different fractions (from 0 to 1) of Hα flux contributed by shocks as indicated. For comparison the theoretical (red curve) and empirical (blue dashed curve in the left-hand panel) upper limits for HII regions are also indicated as well as the separation between AGN and LINER (solid blue line in the central and right-hand panel).

Fig 3 – Star formation rate across SOS 114372 derived from the Hα flux corrected for dust attenuation. The red contours show the isophotes of the flux at 24μm flux. The scale for the (logarithm of) SFR is at the bottom.

We mapped the star formation rate (SFR) across SOS 114372 from the surface distribution of the Hα flux. Intense star formation is found in the disc spanning from 0.01 to 0.34M_☉ yr⁻¹ arcsec⁻² with the global maximum located in the galaxy centre. There is also a notable local maximum in the SW region of the disc at∼12 kpc from the centre, with SFR = 0.2M_☉ yr⁻¹ arcsec⁻².

The integrated Hα-derived SFR of SOS 114372 amounts to 7.2 ± 2.2 M_☉ yr⁻¹. The error takes into account the uncertainties related to the flux and attenuation measurements added to a 30 per cent uncertainty due to different calibrations of Hα as a SFR indicator (Kennicutt 1998). From the difference between raw and attenuation-corrected Hα fluxes we can estimate the obscured SFR as 5.3 ± 1.6 M_☉ yr⁻¹. This obscured SFR is also measured by the IR dust emission turning out to be 6.65+1.99 −1.19 M_☉ yr⁻¹. The two independent estimates of the obscured SFR are therefore fully consistent.

Go to the paper (Merluzzi et al. 2012) or to the presentation (Merluzzi et al. 2012) at the Monash Centre for Astrophysics (Melbourne, Australia, May 2012).

Our multi-band survey of the Shapley supercluster core revealed a number of galaxies in the process of being transformed or in interaction with the environment. A sample of these was observed with WiFeS.
Within this sample, SOS 114372 (identification from the Shapley Optical Survey) is the most IR-luminous and the most actively star-forming (~8 M☉ yr-1) was observed with two pointings of WiFeS, as indicated in Fig. 1.

Fig. 1 – BRK composite image of SOS 114372. The two white rectangles denote the WiFeS pointings: #1 centred on the galaxy, #2 covering the NE side. The two galaxies to the NE are supercluster members: SOS 114493 (z = 0.05) and SOS 112393 (z = 0.046). SWof SOS 114372 is a faint galaxy, SOS 115228, without redshift measurement.
The orientation is indicated and adopted for all the following galaxy maps. The white bars indicating the orientation correspond to a length of 15 kpc. Channels R, G, B are assigned to K, R, B bands, respectively. The nuclear region appears to be highly obscured by dust.

The first result of WiFeS was to reveal the existence of extraplanar ionized gas extending ~ 13 kpc NW of the disc of the galaxy, as shown in Fig. 2.

Fig. 2 – Spatial distribution of the flux of the Hα emission line measured by WiFeS (contours) superimposed on the K-band image of SOS 114372. Note the extension of the ionized gas in the NW (top-left) direction, well beyond the extension of the disc.

The kinematics of the ionized gas, shown in Fig. 3, although characterised an overall rotation in the disc, is clearly perturbed, as is easily seen by the asymmetric iso-velocity contours (left panel). The velocity field of the extra-disc gas appears to be dominated by the rotational motion characterising of the inner disc.
The velocity dispersion has maxima in the extraplanar gas and in a triangular area SE from the centre.
The stellar kinematics, on the other hand, appears symmetric with respect to the centre, with the typical features of a barred galaxy as SOS 114372 is.
Perturbations affect therefore only the gas.

Fig 3 – Kinematics of ionized gas. Left: radial velocity, right: velocity dispersion. White contorurs describe the brightness distribution of the stellar component. Dashed lines mark the apparent minor and major axes fo the galaxy. Black contours in the velocity field are traced to help inspection of asymmetries. NE is on top-left.
To understand if ram-pressure stripping could explain this complex kinematics, we run N-body hydrodynamical simulations for SOS 114372. The simulations were run with different velocities and angles of the galaxy with respect to the inter-galactic gas.
Some simulations were able to reproduce the observe velocity field remarkably well, as shown in Fig. 4.

Fig 4 – The velocity field produced by the “best” N-body/hydrodynamical simulation (left) is compared to the real data (right). The white arrow is the projection on the plane of the sky of the wind of the intra-cluster medium causing the gas stripping off the galaxy.
WiFeS data, interpreted through simulations, show that the interstellar gas of SOS 114372 is being blown off the disc. The galaxy is moving at a speed of ~ 1400 km/s into the inter-galactic medium embedding the cluster Abell 3558. The disc of the galaxy is inclined at about 45o with repect to the motion of the galaxy, so that the gas of the disc feels a wind inclined ~45o with respect to the rotation axis. The interaction of the inter-galactic gas with the interstellar medium of SOS 114372 started about 50-60 million years ago.

The phyisical conditions of the gas are outlined here.
See here for details regarding the comparison of simulations with observations.

Go to the paper (Merluzzi et al. 2012) or to the presentation at the Monash University (Merluzzi et al. 2012) at the Monash Centre for Astrophysics (Melbourne, Australia, May 2012).

 

 

 

The full exploitation of the power of integral-field spectroscopy and N-body/hydrodynamical simulations necessary to understand the complex phenomenon of ram-pressure stripping requires a careful comparison between the theoretical predictions and the data.
We developed a method to achieve this goal, based on a two-step analysis.
Observing the simulations
Simulating the observations
The result of simulations consists, at a given time, in the positions and velocities of particles representing the gas (simulations give a lot of quantities, we will limit here on just position and velocity).
The two steps are the following:
Whatever the reference frame of simulations is, we must transform it to the observer’s reference frame, in other words, we must observe the simulations from the right point of view. This is accomplished by step 1.
Then, we must take into account how our instrument would transfer the information carried by the particles to us, i.e. we have to simulate our observing conditions. This is accomplished by step 2.

 

1 – Observing the simulations

 

The geometry of a simulation of ram-pressure stripping can be described as in the figure below.
XYZ is a cartesian coordinate system, with the disc of the galaxy in the XY plane.
The circles in the figure are the intersection of a unit sphere with the XY plane (left) and with the YZ plane (right).
The small black dot on the unit sphere is the direction of the observer, while the dashed line represents the line of sight. In this way, the X axis is along the apparent major axis of the galaxy (and the Y axis along the minor one).
Notice that two viewing situations, differing by a rotation of 180o around the X-axis, are in principle possible. In one case (case A) the wind points towards Z > 0, while in the other (case B) the wind points towards Z < 0.
Let β be the angle between the inter-galactic (IGM) wind and the Z axis, and φ the angle between the velocity of the galaxy and the observer .
If Vg is the spatial velocity of the galaxy and Vlos is its line-of-sight velocity, then
cos φ=Vlos/ |Vg|. This is equivalent to require that the wind velocity makes an angle φ with the line of sight in the direction of the observer.
The blue circles are the intersection of the cone β=constant (let us call it ‘wind-cone’) with the unit sphere.
The upper panels correspond to β=25o, and the lower to β=45o.
The black/gray circles are instead the cones φ=const (‘φ-cone’). In both figures, the greatest black/gray circle corresponds to φ=70o. The gray portion of the circle is the one below the XY plane with respect to the reader.
The small gray circle in the upper-left figure corresponds to φ=40o.

In the upper-left panel, the two black lines are the projection of the semiplanes passing through the Z-axis and the intersection points of the cones. They are at angles θ1 and θ2 (−32o, 212o in this example) with the X-axis. Rotating the galaxy by one of these two angles, moves the wind direction at the right angle with respect to the line of sight in which it projects into Vlos. In this case there are therefore two possible directions from which we can ‘observe’ our model. In both views, the wind is directed as the red arrow in the upper-right panel, which corresponds to case A. Since the wind cone does not intersect the φ-cone in the Z < 0 hemisphere, no other viewing angles are possible.

In the bottom panels, the wind-cone intersects the φ-cone in both hemispheres. There are two intersections in the Z > 0 hemisphere (black lines) and two for Z < 0 (gray lines), defining in total four angles θ1, …,θ4. Rotating the model by θ1 or θ2 allows us, as before, to see the model from the correct point of view. The red arrows shows the projection of the wind on the YZ plane.
But in this case the model can also be rotated around the Z-axis by θ3 or θ4.
The second case corresponds to the cyan wind vector in the bottom-right panel and represents an example of case B. To reach this configuration, the model should be first rotated by 180o around the X-axis.

The coordinate transformations needed in the above process are the following. Let i be the inclination angle of the galaxy (plane of the sky to galaxy disc). First we move the symmetry axis of the galaxy into the Z-axis with a clockwise rotation of β around the Y-axis. In this way, the wind lies on the (X, Z) plane. The second step is to move the wind vector in a direction in which the wind speed projects into Vg. It can be shown that this is achieved by a rotation around the Z-axis by an angle θ given by.

where k = −1 for case A and k = 1 for case B. For each k this equation has two solutions: θ1= arccos(cos θ) and θ2 = π − θ1. As a result, when only case A is allowed, there are two possible viewing angles (or ‘projections’), while when both cases A and B are allowed, the possible projections are four.

Other transformations are possible that do not change the physics of the models, but given that we fixed the position of the line of sight in the YZ plane, the only remaining meaningful transformation is a reflection with respect to that plane. This corresponds to an inversion of the X-axis and to the change of sign of the corresponding velocity component of the model particles. This transformation makes the projection of the models on the sky to reflect with respect to the apparent major axis, and may become useful to match the models to the observed orientation of the galaxy, when this is not achievable with a simple rotation.

 

2 – Simulating the observations

 

The end-product of the previous process is a matrix (xi , yi , Vi ; i = 1, … , Nparticles), where (x, y) are coordinates in the plane of the sky (x is along the apparent major axis), y is normal to x, V is the radial velocity and Nparticles is the number of particles in the simulation.
To reproduce the effect of the seeing we substituted each particle by a Gaussian distribution with the FWHM equal to the seeing. In each surface element of the plane of the sky (see below), we computed the surface density of particles Σ as the sum of the contributions from the Gaussians and the radial velocity as the weighted average of the individual velocities.
To speed up the computation, we truncated the contribution of each particle at 10σGauss. This computation was performed on a grid of points spaced by 0.1 arcsec on the sky, which we verified to offer a reasonable compromise between accuracy and computation speed. The points in the plane of the sky were eventually sampled in 1 × 1 arcsec2 bins, matching the spatial binning of the data, and rotated to the position angle of the major axis of our galaxy.

​

Fig. 1 –  Example of simulated data (left) and the same data processed as explained in Section 2 (right). The instes in the left panel show the orientation of the IGM wind in the three coordinate planes.

Go to the paper (Merluzzi et al. 2012) .

Our mid-infrared Spitzer data of the Shapley supercluster core revealed the existence of a number of galaxies with extremely “cold” mid-infrared colours, as shown in Fig. 1.

Fig. 1 – Mid-infrared ‘colour’ f70μm/f24μm versus the f70μm flux of the Shapley supercluster galaxies. Each symbol is coloured according to its f24μm/fK flux ratio, as indicated by the coloured band of Fig. 1. The magenta curve shows the predicted  f70μm/f24μm colour evolution of the luminosity-dependent IR SEDs of Rieke et al. (2009). The red scale shows the expected MIR colours of the Dale & Helou (2002) models for α= 0.1–4.

These values of f70/f24 appear inconsistent with any of the spectral energy distributions expected for star-forming galaxies or AGN. In total, we identify 23 ‘70μm-excess’ galaxies having  f70μm/f24μm > 25, or 14 ± 3 per cent of the 70μm-detected Shapley supercluster core galaxies.

They are strongly concentrated towards the cores (≲0.5r₅₀₀; see Fig. 2) of the five clusters that make up the SSC, and also appear extremely rare among local field galaxies from the SINGS and SWIRE surveys, confirming them as a cluster-specific phenomenon.

Fig. 2 – Spatial distribution of spectroscopically confirmed 70μm sources across the Shapley supercluster. The symbols are coloured according to their f70μm/f24μm MIR colour from blue (≲7) to orange/red (≳30), while the sizes scale with the K-band luminosity. The shaded region shows the coverage of WFI B-, R-band imaging from the SOS (Mercurio et al. 2006), from which we determine the morphologies. The grey-scale contours indicate the surface density of R < 21 galaxies (fig. 1 of Haines et al. 2006).

The spectra (Fig. 3) of these galaxies are dominated by old (> 6 Gyr) stellar populations with little or no ongoing star formation.
They are primarily massive galaxies with E/S0 morphologies, as confirmed by subsequent bulge–disc decompositions requiring comparable bulge and disc components to fit their R-band surface brightness profiles (see Fig. 4).

Fig 3 – Optical spectra (shifted to the rest frame of the galaxy) of those 70μm excess supercluster galaxies within the AAOmega sample of Smith et al. (2007) covering the wavelength range 370–510 nm (left) around the Hα emission line (right). Also shown are two SSPs from the MILES stellar population models (Vazdekis et al. 2010) which bracket the properties of the observed spectra. The primary Balmer and forbidden-line indices are indicated by shaded vertical bars.

Fig 4 – R-band images of the IR-cold galaxies taken with the ESO 2.2m telescope.

We put forward two possible causes for these cluster S0s with excess 70-μm emission: (i) assuming the dust-to-gas ratio remains constant, the 70μm excess suggests a reduction in the SF efficiency in these dusty cluster S0s as proposed within the morphological quenching scenario or (ii) a 2–3 times increase in the dust-to-gas ratio or metallicity of the remnant ISM of these galaxies, as predicted by multi zone chemical evolutionary models of galaxies in the process of being ram-pressure stripped or starved. In this latter scenario, the accretion of pristine gas is shut down, and so no longer dilutes the remaining H i gas enriched by metals recycled by stellar mass loss.

Go to the paper (Haines et al. 2011) or to the presentation (Merluzzi et al. 2011) at the Galaxy Evolution and Environment conference (Milano, Italy, November 2011).

We derived the global star formation in the Shapley supercluster from the mid-infrared and far-UV luminosity functions.

The star formation in many of the supercluster galaxies is heavily obscured, with more of the energy from young stars output in the infrared than the ultraviolet. How does this average out over the entire supercluster population? Using the ultraviolet and 24μm-based calibrations of Leroy et al. (2008) we can directly compare in Fig. 1 the observed FUV and 24μm luminosity functions in terms of star-formation rates.

Fig 1. Comparison of the FUV and 24μm luminosity functions in terms of the derived star formation rates (SFR), assuming the UV and MIR components of the SFR calibration of Leroy et al. (2008). The total infrared luminosity in the top axis is estimated from the 24μm fluxes following the calibration of Bai et al. (2009).

For the galaxies with the highest SFRs, the estimates obtained via the mid-infrared are ~10x higher than those obtained from the uncorrected far-ultraviolet, and that for SFRs>1M_☉ yr^-1, the far-UV component can be neglected and the 24mum LF can be used as a good estimator of the SFR distribution. At lower SFRs however (<1M_☉ yr^-1), the FUV component cannot be neglected, as it becomes equally important as the infrared component, comprising anywhere between 10–90 per cent of the emission from star-formation in these galaxies.

Over the entire SOS region, we estimate a total supercluster SFR of 327+102-60 M_☉ yr^-1, of which 264+102-60 M_☉ yr^-1 (~80 per cent) is obscured (based on the 24μm calibration) and just 63+/-3 M_☉ yr^-1 (~20 per cent) is unobscured and emitted in the form of ultraviolet continuum. For the mid-infrared contribution we excluded those galaxies for which we assume the 24mum emission comes from evolved stars (see Haines et al. 2010a for details).

We believe the bulk of the mid-infrared emission (and hence our SFR budget) to be due to star formation, rather than AGN, whose contribution we estimate as of the order 5-10 per cent (see Haines et al. 2011b).

Go to the paper Haines et al. 2011a.

One of the key aims of the infrared surveys of galaxies in group/cluster environments is to obtain complete censuses of star-formation in dense environments, and to establish whether there is a population hidden from previous UV/optical surveys with significant obscured SFRs sufficiently large to build up an S0 bulge in the time-scale required to transform the star-forming spirals seen at z~0.4 clusters into the passive S0s that have empirically replaced them in local clusters. In the massive cluster Cl0024+16 at z~0.4 Geach et al. (2009) identified a population of heavily obscured LIRGs with 24μm-derived SFRs of 30-60 M_☉ yr^-1 and Spitzer/IRS mid-infrared spectra consistent with nuclear starbursts. We find no such galaxies with SFRs in the range 20-40 M_☉ yr^-1 (converting to our Kroupa IMF) within the Shapley supercluster, comparable to those found by Geach et al. (2009). All the supercluster members are observed to have SFRs (including both obscured and unobscured components) in the range 0.1-13 M_☉ yr^-1, and specific-SFRs implying mass-doubling time-scales >1 Gyr. This limit to relatively modest SFRs (<15 M_☉  yr^-1) has been found for galaxies in all local clusters analysed to date in the mid-infrared out to z~0.2. A number of mechanisms have been proposed to transform infalling spiral galaxies into passive S0s which include intermediate starburst and possible subsequent post-starburst phases.

We identify starburst galaxies in the Shapley supercluster core via three methods: (i) their location above the specific-SFR-M relation or equivalently having flux ratio f₂₄/f_K>5 (see the page on bimodality);

Fig. 1 Relationship between specific-SFR and stellar mass for the supercluster galaxies. The SFR is the sum of the SFR(FUV) and SFR(24mum) components when galaxies are detected at 24mum (filled symbols) and SFR(FUV) only otherwise (open symbols). Stellar masses are derived from the BRK photometry as described in Merluzzi et al. 2010. The filled symbols are coloured according to f_24/f_K (see bimodality). The black dot-dashed line indicates the SFR corresponding to our 24mum completeness limit. The green solid and dashed-lines indicate the SSFR-M relation and scatter of local (0.015<z</=0.1) blue star-forming SDSS galaxies of Elbaz et al. (2007), while the blue line indicates the corresponding relation of Schiminovich et al. (2007).

(ii) their location along the starbust IRX-beta relation of Kong et al. 2004, (blue shaded region of Fig. 1);

Fig. 2. The IRX-beta plot for galaxies in the Shapley supercluster. Symbols are colour coded (richiamo bimodality) and are of sizes which scale with their K-band luminosity. Galaxies with f_24>5 mJy for which the 24mum emission remains unresolved are ringed. The magenta dashed line indicates the Kong et al. 2004) starburst curve. The blue shaded region indicates the selection criteria used to identify our starburst galaxies. The solid lines indicate models of ageing instantaneous burst populations convolved with the starburst opacity curve for increasing amounts of dust attenuation from Fig. 10 of Calzetti et al. (2005) of 2 (blue line), 12 (green line) and 300 Myr old (red line). The dot-dashed line marked “2pops” shows the model track for the combination of a 5 Myr old instantaneous burst with a 300 Myr old one. The mass of the 5 Myr old burst is 300 times lower than that of the 300 Myr old one, and its extinction is systematically higher by Delta A_V=0.24 mag; both population models are convolved with the starburst opacity curve.

(iii) having unresolved 24μm emission indicative of a nuclear starburst (indicated by ringed symbols in Fig. 1).

Irrespective of which method we consider, we identify ~15 starburst galaxies across the survey, contributing just 15% of the global SFR within the supercluster core.

The dominant contributor (~85 per cent) to the global SFR budget in the Shapley supercluster is rather spiral galaxies undergoing normal, quiescent star formation across their disks. This star-formation occurs in a manner indistinguishable from the general field population (see Haines et al. 2010 for details), consistent with them being galaxies recently accreted from the field, but yet to encounter the dense ICM or be affected by cluster-related processes.

Go to the paper Haines et al. 2011b.

Fig. 1 The infrared colour f₂₄/ f_K versus f₂₄ for Shapley supercluster galaxies. The histogram on the right shows the distribution of f₂₄/ f_K colours of the 24μm detected supercluster galaxies, revealing a clear bimodality in the infrared colours. The green points are spectroscopic supercluster members while the blue points indicate those objects selected photometrically as supercluster members.

In Figure 1 we show that the infrared colours of the supercluster galaxies to be bimodally distributed in f₂₄/ f_K, with two peaks well separated by a clear gap where few galaxies are located. Given the well known bimodality in galaxy colours, morphologies and spectroscopic properties in which galaxies can be broadly split into blue, star-forming spirals and red, passively-evolving ellipticals, it is natural to ascribe the observed bimodality in infrared colour, which can be considered a proxy for specific-SFR, as another manifestation of this division. In this case, we would expect the f₂₄/ f_K flux ratio to correlate strongly with morphology or other indicators of star-formation history such as UV-optical colours or spectral indices including EW(Hα).

Fig. 2 The infrared colour f₂₄/ f_K versus f₂₄ flux of galaxies from the SINGS sample (Kennicutt et al. 2003), where the f₂₄ flux is based upon the galaxy being shifted to the distance of the Shapley supercluster. The colour of the symbol indicates its morphological class from red (E) to blue (Im), as indicated.

In Fig. 2 we reproduce the diagram of Fig. 1 for the galaxies in the SINGS sample. Although the actual bimodality is not quite reproduced, we can now identify the sequence at f₂₄/ f_K ~0.05 with the passive E/S0 population, while the sequence at f₂₄/ f_K~2 corresponds broadly to the mass sequence of Sb-Sd spirals and irregulars, albeit along with a number of earlier-type spirals and dusty S0s. Intermediate between the two main sequences we find primarily Sab spirals.

Fig.3 Correlation between the f₂₄/ f_K MIR-K-band flux ratio with the FUV-R UV-optical colour (top panel) and EW(Hα) obtained from the AAOmega spectroscopy (bottom panel). The colour of each symbol indicates the corresponding spectroscopic classification: (red) passive; (green) AGN; (cyan) composite; (magenta) unknown; (blue) star-forming.

It is immediately apparent from Fig. 3 (top panel) that we can robustly split galaxies into star-forming or passive populations about a flux ratio f₂₄/ f_K~0.3. We can thus identify the two sequences seen in Fig. 1 as: (i) the blue cloud of star-forming galaxies with f₂₄/ f_K~0.5-10; and (ii) the red sequence of passively-evolving galaxies having f₂₄/ f_K~0.02-0.3. In Fig. 3(top panel) the strong correlation between the FUV-R colour and f₂₄/ f_K is shown: galaxies in the infrared passive sequense ( f₂₄/ f_K<0.15) all lie on the UV-optical red sequence (FUV-R~7), galaxies of the infrared star-forming sequence ( f₂₄/ f_K~2) lie within the UV-optical blue cloud (FUV-R<5).

Go to the paper Haines et al. 2011a.

We studied the optical LFs (B and R bands) of the Shapley supercluster core (SSC, Mercurio et al. 2006) with the aim to investigate the effects of the environment on the photometric properties of galaxies. We derived the galaxy LFs in three regions selected according to the local galaxy density, and found a marked luminosity segregation, in the sense that the LF faint-end is different at more than 3σ confidence level in regions with different densities. In addition, the LFs of the red and blue galaxy population show very different behaviours: while the red sequence counts are very similar to those obtained for the global galaxy population (showing dips at about M*+2 and a faint-end upturns), the blue galaxy LFs are well described by a single Shechter function and do not vary with the density. Such large environmentally dependent deviations from a single Schechter function are difficult to produce solely within galaxy merging and suffocation scenarios. Instead, the optical data support the idea that mechanisms related to the cluster environment, such as harassment or ram-pressure stripping, shape the galaxy LFs by terminating star formation and producing mass-loss in galaxies at M*+2, a magnitude range where the blue late-type spirals used to dominate cluster population, but are now absent.

Figure 1: optical Luminosity Functions. Left panel: R-band galaxy LF in the high- (ρ>1.5, black), intermediate- (1<ρ≤1.5, red) and low-density (0.5<ρ≤1, blue) SSC environments. Continuous lines represent the Gaussian + Schechter best fit for intermediate- and low-density environments and Schechter best fit for the high density environment. Right panel: the 1, 2, and 3σ confidence levels for the corresponding Schechter parameters.

 

 

Go to the paper Mercurio et al. 2006.

The optical luminosities of galaxies are strongly affected by the presence of young stellar populations and dust, which can bias the optical LFs, making it difficult to reliably interpret the data. The near-infrared (NIR) instead is much less sensitive to the effects of dust or star formation, and hence the NIR (e.g. K-band) LF can be considered a reliable estimator of the underlying stellar mass function (Bell & de Jong 2001).

We studied the near-infrared luminosity and stellar mass functions of galaxies in the SSC (Merluzzi et al. 2010). The K-band survey, together with the SOS and a subsample (~650 galaxies) of spectroscopically confirmed supercluster members, allowed us to investigate the supercluster galaxy population down to M_K*+6 and M=10^8.75M_☉. As for the optical LF we investigated the effect of environment by deriving the LF in three regions selected according to the local galaxy density, and observed a significant (2σ) increase in the faint-end slope going from the high- (α=-1.33) to the low-density (α=-1.49) environments, while a faint-end upturn at M_K>-21 becomes increasingly apparent in the lower density regions.

Figure 1. Left panel: K-band galaxy LF in the high- (black), intermediate- (red) and low-density (blue) SSC environments. Continuous lines represent the Schechter best fits. Right panel: the 1, 2, and 3σ confidence levels for the corresponding Schechter parameters.

The galaxy stellar mass function (SMF) of supercluster galaxies is characterised by an excess of massive galaxies that are associated to the cluster BCGs. While the value of M* depends on environment increasing by 0.2 dex from low- to high-density regions, the slope of the galaxy SMF does not vary with the environment.

Figure 2. The mass function of galaxies in the three cluster regions corresponding to high- (left) , intermediate- (centre) and low-density (right) environments. In the left, central and right panel the continuous line represents the fit to the data. In each panel the best fit value of α and log₁₀ M* are reported.

By comparing our findings with cosmological simulations, we conclude that the environmental dependences of the LF are not primarily due to differences in the merging histories, but to processes which are not treated in the semi-analytical models, such as tidal stripping or harassment.

Figure 3. Composite K-band LFs for the 20 most massive clusters in the Millennium simulation, based on the semi-analytic models of Bower et al. 2006 (thick lines) and Font et al. 2008 (thin lines), in bins of projected cluster-centric radius. The four different coloured curves correspond to r<0.5 r₅₀₀ (red solid lines), 0.5<r<1.0 r₅₀₀ (black dashed lines), 1.0<r<2.0 r₅₀₀ (green dot-dashed lines) and r>2.0 r₅₀₀ (blue dot-dot-dashed lines).

In field regions the SMF shows a sharp upturn below M=10⁹ M_☉, close to our mass limit, suggesting that the upturns seen in our K-band LFs, but not in the SMF, are due to this dwarf population. The environmental variations seen in the faint-end of the K-band LF suggests that these dwarf galaxies, which are easier to strip than their more massive counterparts, are affected by tidal/gas stripping upon entering the supercluster environment.

Go to the paper Merluzzi et al. 2010.

The ultraviolet (∼2000Å) provides a measure of the recent SFR over the last 10⁸ yr in galaxies, being dominated by emission from young stars of intermediate masses (2-5 M_☉; Boselli et al. 2001), and hence the ultraviolet LF represents a useful tool to quantify the effects of the cluster environment on star formation.

We estimated the UV LFs of the SSC using two different methods for the background subtraction. The FUV (blue) and NUV (red) galaxy luminosity functions obtained using these methods are shown in Fig.1. We find that the Schechter function well describes the FUV and NUV data.

Figure 1.The FUV (blue) and NUV (red) LFs for the whole 2.8334 deg² region covered by both GALEX and WFCAM K-band imaging obtained using the two different methods for background subtraction: (i) the statistical subtraction of field galaxies method (open circles) and (ii) UV-optical colour selection (solid circles). The dashed and solid lines indicate the best-fitting Schechter functions to the data using the two methods. The error bars include Poisson uncertainties as well as uncertainties due to selection effects and completeness correction (see Haines et al. 2010a for details).

Figure 2 examines the relative contributions of passive (NUV-R≥4.5) and star-forming (NUV-R≤ 4.5) galaxies to the far-UV luminosity function.

Figure 2. The FUV LFs for the passive (NUV-R>4.5; red symbols) and star-forming (NUV-R<4.5; blue symbols) supercluster galaxy sub-populations. The solid lines indicate the best-fitting Schechter functions to the data.

We obtain best-fit Schechter functions for the star-forming supercluster galaxy populations consistent (within the 1σ errors) with those obtained by Wyder et al. (2005) and Budavari et al. (2005) for local field galaxies The contribution of the passive galaxies to the far-ultraviolet LF is negligible at the bright end (M_FUV<-17), but becomes increasingly important at fainter magnitudes.

The pile-up of quiescent galaxies in the last three magnitude bins (M_FUV>-15) appears the primary cause of the discrepancy in the faint-end slopes between cluster and field galaxies. This effect is seen also in the Coma cluster ultraviolet LF (Cortese et al. 2008), and also at optical wavelengths (Mercurio et al. 2006). A second contributing factor to the steeper observed cluster UV LFs could be simply the steeper stellar mass functions (or K-band LFs) of cluster galaxies, as observed by Merluzzi et al. (2010) for this same Shapley supercluster core region.

Given that the FUV emission from the “passive” galaxies is unlikely to be due to star-formation (O’Coconnell 1999; Dorman et al. 2001), the relative consistency between the FUV LFs of cluster star-forming galaxies and the field suggests little environmental dependence in the unobscured SFRs of star-forming galaxies.

Go to the paper Haines et al. 2010a.

 

In many galaxies more than 90% of the UV photons are absorbed by dust, corresponding to A(FUV)>2.5 and hence potentially significantly biasing the ultraviolet LF. This energy absorbed by the dust is however reprocessed as thermal radiation in the mid/far-infrared (8-1000 μm), and so observations at these wavelengths allow us to quantify this energy and thus infer the amount of obscured star formation (e.g. Kennicutt 1998; Calzetti et al. 2007, Rieke et al. 2009; Kennicutt et al. 2009).

We calculate the 24μm luminosity function for the 2.25 deg² region covered by 24μm, optical and GALEX UV data (see Haines et al. for details). The resulting 24μm luminosity function, shown by the solid green points in Fig.1, covers a factor 300 in luminosity, extending down to our survey limit of 0.35 mJy.

We obtain a best-fit Schechter function of L₂₄μm=24.83 (+3.35 −4.65) mJy and α(24μm)=−1.425 (+0.035 −0.040), shown by the curve in Fig.1. Our LFs are fully consistent with both those obtained by Bai et al. (2006) for Coma and Bai et al. (2009)} for Abell 3266.

Figure 1. 24μm luminosity function for the Shapley supercluster (solid green symbols). The contribution due to spectroscopically confirmed supercluster members is indicated by open symbols. Blue squares represent the field 24\mum LF of Marleau et al. (2007) for 0<z<0.25 star-forming galaxies in the Spitzer First Look Survey. Along the top axis we show the corresponding total infrared luminosity (L_IR[8-1000 μm], see Haines et al. 2010a for details).

Since the shape of the 24μm LF of Shapley supercluster galaxies is fully consistent with that obtained for local field galaxies, our findings support the assertion of Bai et al. (2009) that there is no environmental dependence of the shape of the 24μm luminosity function.

The Spitzer/MIPS 70 μm filter, being closer to the far-infrared peak affords more reliable L_IR estimates. Figure 2 shows the corresponding 70 μm luminosity function of Shapley supercluster galaxies.

We fit the 70 μm LF by a single Schechter function as shown by the green curve, obtaining best-fit parameters fully consistent with those obtained at 24 μm, albeit with larger uncertainties due to the relatively shallow depth. We compare the supercluster 70 μm with the 60 μm local field galaxy luminosity function (Wang & Rowan-Robinson 2010; Takeuchi et al. 2003). Again, we find no significant difference between the 70 μm luminosity function of Shapley supercluster galaxies and that obtained for local field galaxies.

This represents the first measurement of the 70 μm galaxy LF of a local cluster with Spitzer, and should prove a useful local benchmark to follow the evolution in the infrared properties of cluster galaxies with ongoing and future surveys (e.g. Haines et al. 2009a,b ; Haines et al. 2010; Smith et al. 2010 ; Braglia et al. 2010 ; Rawle et al. 2010 ; Chung et al. 2010) with Spitzer and particularly Herschel whose PACS instrument covers rest-frame 70 μm beyond z~1.

Figure 2. 70 μm luminosity function for the Shapley supercluster (solid green symbols). The contribution due to spectroscopically confirmed supercluster members is indicated by open symbols. The best fitting Schechter function is indicated by the solid green curve and the associated parameters and uncertainties indicated. The blue squares indicate the local field luminosity function of IRAS galaxies from Wang & Rowan-Robinson (2010), while the blue dashed curve indicates the analytic form of the local field IR luminosity function of Takeuchi et al. (2003), both obtained from the PSCz survey.

A probable significant consequence of this apparent complete agreement between the local cluster and field galaxy infrared LFs, is that the bulk of star-forming galaxies that make up the observed cluster infrared LFs have been recently accreted from the field and haven’t had their star formation activity significantly affected by the cluster environment yet, based on a similar argument to that used by Balogh et al. (2004) regarding the lack of environmental dependence seen for the distribution in EW(Hα) (see Haines et al. 2010a for the complete discussion).

Go to the paper Haines et al. 2010a.

Early-type galaxies obey a scaling relation between effective (“half-light”) radius r_e, mean surface brightness within r_e <I_e>, and central velocity dispersion σ₀. In the 2-D space defined by (log(r_e), log(<I_e>), log(σ₀), early type galaxies populate a tight plane known as the Fundamental Plane, expressed by

log r_e = α log σ₀ + β log <I_e> + γ

Figure 1. “Edge-on” view of the FP of the Shapley early-type galaxies with magnitude R < 18.

Once accounted for the measurement errors, there is still a residual scatter around the FP, which must be intrinsic to the properties of the galaxies. The origins of this intrinsic scatter have been long investigated. The deviation of early-type galaxies from the FP was found to be related to the properties of their stellar populations, such as the metal content, the colour, and the age. (for a review see Gargiulo et al. 2009)

The present work is based on a sample of 141 early-type galaxies of the Shapley supercluster core. The spectroscopic data come from the AAOmega survey of the SSC by Smith et al (2007). The photometric data are those of the SOS.
One key feature of this survey is its coverage of low-mass galaxies down to σ $sim;50km s^-1.

Among age, metallicity, and α-enhancement, the last property of stellar populations is the most strongly correlated with the residuals to the FP, as shown in the next figure. Galaxies with higher α/Fe have smaller effective radii. Age is also anti-correlated with the residuals to the FP, in the sense that older galaxies are more compact.

Figure 2. Correlation of the stellar population properties with the residual to the the FP. Left to right: age, metallicity, and α-enhancement. The residual to the FP is measured in the direction of effective radii.

The dependence of α/Fe on the residuals to the FP is illustrated in the next figure.

 

Below we show, for each of the galaxies in the sample currently observed with WiFeS, the R-band (λ~700nm) image on the left panel and the same image with overimposed the velocity field of the ionized gas on the right.

Even a direct visual comparison between the image on the left and the velocity field on the right may give a first rough idea on what is going on in each system. As an example, in the galaxy below the gas is escaping from the main galaxy body both in the upper-left and lower-right directions. The gas is instead escaping towards one precise direction (right) from the galaxy in the second row.

Below we collect the kinematic data (radial velocity and velocity dispersion) of a number of galaxies observed with WiFeS. The kinematics is derived from the Hα emission line and thus represents the motion of the ionized gas.

In all plots, the black contours show the distribution of the stars.

The left panels show the radial velocity of the gas, namely the velocity with respect to the observer. The right panels show the velocity dispersion, which is mostly a measure of the turbulent motions of the gas within each resolution element (the small squares) of the figure.

We can identify galaxies with part of their gas being swept away (figures 1, 2, 3, 4, 6, 8) and galaxies with anomalous (very fast or very asymmetric) rotation (figures 7 and 9). Galaxies  5, 10 and 11 host active galactic nuclei.

 

Depths of the survey
(AB magnitudes in VST-SDSS filters)

detection limit at S/N=5 within 3” aperture: r=24.6, u=25.0, g=25.0, i=23.8, z=23.0

limit for star/galaxy separation: r=23.5 (L*+8.5)
These depths will allow to

obtain photometric redshifts
analyze the internal structure of galaxies
study the internal distribution of stellar populations
derive morphological classification
estimate unobscured star-formation rates
down to L*+6, well into the dwarf regime.

100 VST GTO hours are allocated for this project, that will be carried out in four years starting in 2012.

The survey started in February 2012.

The status of the observations, updated to February 2013, is shown in the following figure.

Data are reduced with the VST-Tube pipeline (Grado et al. 2012 Mem. SAIt Supp. 19, 362).

Catalogues of galaxies and stars have been extracted.

The following snapshot were extracted from different VST pointings. The black-and-white images are in the g band (1400 s integration, ~0.5” FWHM seeing) where not indicated, while the colour images are composite from g, r and i bands. The images are shown to give a feeling of the high quality of VST-OmegaCAM data.

 

 

P. Merluzzi (P.I., INAF-OACN)

G. Busarello, A. Mercurio, A. Rifatto (INAF-OACN)

C. P. Haines (Steward Observatory, Arizona)

M. A. Dopita, F. Vogt (Australian National University)

K. Pimbblet (Monash University, Melbourne)

S. Raychaudhury, G. P. Smith (University of Birmingham)

R. J. Smith (University of Durham)