Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.eso.org/~hkuntsch/papers/ApJ_705_199.pdf Äàòà èçìåíåíèÿ: Fri Mar 5 13:17:07 2010 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 06:05:17 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ngc 4261

The Astrophysical Journal, 705:L199­L203, 2009 November 10
C

doi:10.1088/0004-637X/705/2/L199

2009. The American Astronomical Society. All rights reserved. Printed in the U.S.A.

EVIDENCE FOR INTERMEDIATE-AGE STELLAR POPULATIONS IN EARLY-TYPE GALAXIES FROM K-BAND SPECTROSCOPY
´ ´ ´ ´ E. Marmol-Queralto1 , N. Cardiel1 , P. Sanchez-Blazquez2 , S. C. Trager3 , R. F. Peletier3 , H. Kuntschner4 , D. R. Silva5 , A. J. Cenarro2 , A. Vazdekis2 , and J. Gorgas1
1

4

Dpto. de Astrof´ isica, Universidad Complutense de Madrid, E28040, Madrid, Spain; emq@astrax.fis.ucm.es 2 Instituto de Astrof´ isica de Canarias. c/V´ Lactea S/N, E38205, La Laguna, Tenerife, Spain ia ´ 3 Kapteyn Astronomical Institute, University of Groningen, Postbus 800,9700 Av Groningen, Netherlands Space Telescope--European Coordinating Facility, Karl-Schwarzschild-Str. 2, D-85748 Garching, Germany 5 National Optical Astronomy Observatory, 950 North Cherry Avenue, Tucson, AZ, 85719, USA Received 2009 June 10; accepted 2009 September 28; published 2009 October 26

ABSTRACT The study of stellar populations in early-type galaxies in different environments is a powerful tool for constraining their star formation histories. This study has been traditionally restricted to the optical range, where dwarfs around the turn-off and stars at the base of the red giant branch dominate the integrated light at all ages. The near-infrared spectral range is especially interesting since in the presence of an intermediate-age population, asymptotic giant branch stars are the main contributors. In this Letter, we measure the near-infrared indices Na i and DCO for a sample of 12 early-type galaxies in low-density environments and compare them with the Fornax galaxy sample presented by Silva et al.. The analysis of these indices in combination with Lick/IDS indices in the optical range reveals that (1) the Na i index is a metallicity indicator as good as C4668 in the optical range, and (2) DCO is a tracer of intermediate-age stellar populations. We find that low-mass galaxies in low-density environments show higher Na i and DCO than those located in the Fornax cluster, which points toward a late stage of star formation for the galaxies in less dense environments, in agreement with results from other studies using independent methods. Key words: galaxies: clusters: general ­ galaxies: elliptical and lenticular, cD ­ galaxies: evolution ­ galaxies: formation ­ galaxies: stellar content

1. INTRODUCTION For the majority of galaxies in the universe, the analysis of the stellar content is restricted to the integrated light of their stellar populations, where a small percentage of luminous stars can change the spectra dramatically and bias the final results. A way to overcome this bias is to use a multiwavelength approach, since the fractional contribution from different stars varies along the spectral range. However, most of the observational effort has focused on the optical range, mainly due to the scarcity of appropriate instrumentation in other spectral regions. Especially interesting is the study of the near-infrared (near-IR) window, where red supergiants in young populations (<1Gyr), asymptotic giant branch (AGB) stars in intermediate-age populations (1­2 Gyr), and the tip of the red giant branch (RGB) stars in the near-IR are the main contributors, as opposed to turn-off dwarfs and stars at the lower part of the RGB in the optical. Thus, near-IR spectroscopy may allow a breaking of the degeneracy between multiple stellar populations that plagues optical absorption-line strength studies (e.g., Trager et al. 2000; ´ ´ Sanchez-Blazquez et al. 2006b). In addition, the near-IR wavelength range is less affected by line blanketing effects than other spectral intervals. ´ ´ Marmol-Queralto et al. (2008, hereafter MQ08) presented a detailed study of the CO band at 2.3 m as a function of the basic stellar atmospheric parameters (effective temperature, surface gravity, and metallicity) using a new spectral library of stars in the K band. This work confirmed that this CO absorption band is sensitive to metallicity and it also shows good evidence for increased CO absorption strength in AGB stars compared to non-AGB stars (see Figure 14 in MQ08 and Figure 1 in this Letter). These results confirm the suitability of the CO index for stellar population studies. Furthermore, in this spectral range L199

there is a prominent absorption feature at 2.2 m mainly due to sodium and scandium (Wallace & Hinkle 1996). This feature is very sensitive to the effective temperature of the stars. Once that parameter is fixed, AGB and RGB stars show very similar values (see Figure 1). The study of galaxies in different environments is a powerful tool to understand their evolution and star formation history ´ ´ (e.g., Sanchez-Blazquez et al. 2003, hereafter SB03; Kuntschner ´ ´ et al. 2002; Thomas et al. 2005; Sanchez-Blazquez et al. 2006a; Collobert et al. 2006). In this Letter, we investigate possible differences in their stellar populations comparing the more prominent absorption features in the K band. Only a few detailed studies have been presented in this spectral range after the pioneering photometric work by Frogel et al. (1978) and Aaronson et al. (1978) on early-type galaxies. The first detailed spectroscopic analysis in the K band of these galaxies depending on their environment was presented by Mobasher & James (1996), who used the strength of the first CO band at 2.3 m as an indicator of the presence of AGB stars and, therefore, of an intermediate-age population in elliptical galaxies. They explored the influence of environment finding, initially, a systematic difference between field and cluster galaxies that was later contradicted in James & Mobasher (1999), when the authors enlarged their sample. They also studied the CO strength in Coma cluster galaxies (Mobasher & James 2000), finding variations in the strength of this feature as a function of cluster-centric distance, with more distant galaxies showing deeper CO absorption. Silva et al. (2008, hereafter S08) observed a sample of 11 early-type galaxies (see Table 1) in the Fornax cluster, and analyzed several line-strength indices (Na i,Ca i, Fe i , and CO) in the K band. For those galaxies showing no trace of young or intermediate-age populations, they found a strong correlation


L200

´ ´ MARMOL-QUERALTO ET AL.
Table 1 Measurements for the Early-type Galaxies in This Study Galaxy Type E4-5 E5 E2-3 E6 E-S0 E4 E E1-2 E3 E S0 E S0pec E5 E0 S0 E0 S0a S0 E0 E2 E0 E4 H a (å) 2.03 1.52 1.63 1.60 1.53 3.30 1.50 1.59 1.99 1.42 1.78 1.86 2.39 2.14 1.56 2.93 1.68 1.70 1.63 1.29 1.46 1.62 1.58 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.10 0.05 0.05 0.05 0.09 0.11 0.07 0.05 0.15 0.11 0.12 0.10 0.11 0.09 0.15 0.12 0.12 0.15 0.09 0.16 0.11 0.18 0.08 Mgb (å) 3.83 4.47 4.83 5.19 5.16 2.67 5.39 4.66 4.16 4.94 3.81 3.59 4.01 3.85 5.04 2.62 4.43 4.85 4.32 5.98 4.95 3.97 4.33 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
a

Vol. 705

C4668 (å) 6.70 8.20 8.71 10.01 7.45 5.72 8.97 7.53 8.43 ± ± ± ± ± ± ± ± ± ··· ··· ··· ± ··· ± ± ± ± ± ± ± ± ±

b

Na i (å) 3.46 4.75 5.10 5.68 4.03 4.16 5.48 4.01 4.95 4.61 4.00 3.87 4.61 4.33 4.12 3.50 2.87 4.51 3.32 5.42 4.78 2.86 3.43 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.08 0.12 0.26 0.14 0.14 0.03 0.21 0.14 0.08 0.14 0.11 0.08 0.15 0.15 0.13 0.23 0.19 0.10 0.15 0.23 0.08 0.17 0.17

DCO 1.2028 1.2196 1.2107 1.2314 1.2138 1.2246 1.2212 1.2099 1.2220 1.2115 1.2188 1.2207 1.2258 1.2103 1.2136 1.1952 1.1967 1.2062 1.2083 1.2036 1.2163 1.1982 1.2012 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.0016 0.0025 0.0022 0.0019 0.0019 0.0009 0.0021 0.0021 0.0019 0.0021 0.0027 0.0016 0.0015 0.0005 0.0006 0.0006 0.0006 0.0002 0.0005 0.0022 0.0006 0.0004 0.0004

d (km s-1 ) 59.1 199.5 300.9 184.3 221.2 89.2 305.4 226.7 151.0 249.2 118.2 150.1 226.0 171.0 196.0 70.0 130.8 213.0 169.0 360.0 222.2 128.0 186.0 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 7. 0 20.7 29.4 24.8 23.7 9. 6 22.3 25.7 14.5 21.0 12.6 17.9 8. 0 8. 0 9. 0 9. 0 8. 2 6. 0 8. 0 8. 0 7. 7 6. 0 11.0

Environmente Low density Low density Low density Low density Low density Low density Low density Low density Low density Low density Low density Low density Fornax Fornax Fornax Fornax Fornax Fornax Fornax Fornax Fornax Fornax Fornax

NGC 3605 NGC 3818 NGC 4261 NGC 4564 NGC 4636 NGC 4742 NGC 5796 NGC 5813 NGC 5831 ESO382-G16 ESO446-G49 ESO503-G12 NGC 1316 NGC 1344 NGC 1374 NGC 1375 NGC 1379 NGC 1380 NGC 1381 NGC 1399 NGC 1404 NGC 1419 NGC 1427

0.15 0.09 0.12 0.11 0.15 0.11 0.16 0.12 0.12 0.15 0.14 0.13 0.13 0.10 0.20 0.16 0.15 0.21 0.12 0.24 0.15 0.23 0.11

0.30 0.21 0.21 0.20 0.41 0.20 0.23 0.21 0.22

8.02 7.07 4.93 5.16 8.50 5.99 8.93 8.52 4.97 6.20

0.19 0.24 0.22 0.23 0.29 0.16 0.39 0.24 0.24 0.14

Notes. a H and Mgb indices measured on the optical spectra, extracted within 1/8R , by SB03 and Kuntschner (2000). eff b C4668 index. New measurements on the optical spectra extracted within 1/8R eff by SB03 and original data for Fornax galaxies (Kuntschner 2000). d Central velocity dispersion derived from the near-infrared data by S08 and E. Marmol-Queralto et al. (2009, in preparation). ´ ´ e Following the criteria in Sanchez-Blazquez et al. (2006a) and Kuntschner et al. (2002). ´ ´ Fornax galaxies also observed in this study.

between Na i and CO with velocity dispersion ( ), as well as between those indices and the optical metallicity indicator [MgFe] (Thomas et al. 2003). However, the galaxies with young populations in their sample separate from the global trends. More recently, Cesetti et al. (2009) obtained low-resolution spectra for 14 early-type galaxies and confirmed the relation between the near-IR and the optical features. To unravel possible differences in the stellar content of galaxies in different environments, we analyze a sample of elliptical galaxies located in low-density environments (hereafter field galaxies6 ) in comparison with the Fornax galaxy sample presented by S08. This Letter presents the first results of this spectroscopic study, focusing on the Na i and the CO indices. The rest of the indices, as well as a detailed description of the data ´ reduction, will be presented in a forthcoming paper (E. Marmol´ Queralto et al. 2009, in preparation). In Section 2 we describe the data used in this study, while the results and discussion are presented in Sections 3 and 4, respectively. Finally, in Section 5 we present the main conclusions derived from this work. 2. THE DATA Long-slit spectra for 12 early-type field galaxies (see Table 1) ´ ´ from the samples of Sanchez-Blazquez et al. (2006a) and Kuntschner et al. (2002) plus 2 Fornax galaxies from S08 (for comparison purposes) were obtained with the ISAAC near-IR
6

This sample includes two galaxies in the Virgo cluster. SB03 showed that there is no difference in their optical indices when comparing them with more isolated galaxies, and for that reason they have been included here.

imaging spectrometer mounted at 8.2 m UT1/Antu telescope at Cerro de Paranal (Chile) with exactly the same instrumental configuration employed by S08. Thus, both samples can be readily compared. The field galaxy sample was chosen to cover a range in similar to that spanned by S08's Fornax sample. The spectrometer was used in medium-resolution mode (SWS1 ­ MR), providing a resolution of 7.1 å at 2.3 m (FWHM) with a reciprocal dispersion of 1.21 å pixel-1 for a 120 â 1 slit. The central wavelength was chosen for each galaxy to include both Na i features and the first CO band (2.20 m and 2.29 m at rest frame, respectively). For each object, several exposures along the slit (A and B positions) were taken for sky subtraction, as is the usual practice for infrared observations. The slit was oriented along the minor axis of the galaxies when possible. After each galaxy set, a nearby A-type star was observed for relative flux calibration and telluric correction. Finally, halogen lamps (on and off) and arc lamps were observed for flatfielding, and C-distortion correction and wavelength calibration, respectively. We carried out a standard data reduction in the infrared using REDUCEME (Cardiel 1999). The details of the process will be presented in a forthcoming paper. Galaxy spectra were extracted within a radius corresponding to 1/8Reff at the observed position angle for each galaxy, where Reff is the effective radius. We measured the Na i index (Frogel et al. 2001) and the DCO index (MQ08). The Na i measurements were corrected for broadening effects using the -correction computed by S08, while the DCO index is almost insensitive to this effect (see MQ08). In Table 1 we list these measurements for all the galaxies, together with the central velocity dispersion ( )


No. 2, 2009

INTERMEDIATE-AGE STELLAR POPULATIONS IN EARLY-TYPE GALAXIES

L201

Figure 1. Na i and DCO indices as a function of = 5040/Teff for the subsample of the K-band stellar library of MQ08 with the highest quality spectra. The meaning of symbols and colors is explained in the inset. The AGB stars have higher DCO values than giants, while this behavior is not observed in the Na i index.

computed for each galaxy from these infrared spectra. Previous optical work is available for all galaxies. The optical Lick/ IDS indices Mgb, C4668, and H were measured on the spectra extracted within a radius corresponding to 1/8Reff in the original ´ ´ ´ data by Sanchez-Blazquez et al. (2003; see also Sanchez´ Blazquez et al. 2006a) for the field galaxies. For Fornax cluster galaxies, Mgb and H were taken from S08 (measured on the spectra extracted within 1/8Reff ) and C4668 from Kuntschner (2000; original data). All these optical indices have also been included in Table 1. 3. RESULTS Since there are no reliable stellar population models available describing the behavior of the near-IR indices used in this study, we will interpret the measured features by comparing them with other, better understood, indices. Due to the very strong correlation between metallicity and for Fornax elliptical galaxies (Kuntschner 2000), we use as a "proxy" for global metallicity in these galaxies. As a "proxy" for the mean age, we will use the H index, which is an indicator of the temperature of the turn-off and, therefore, mainly sensitive to the presence of young stellar populations (between 2 â 108 and 109 yr; e.g. Allard et al. 2006). Figure 2 shows the behavior of different

Figure 2. Left column: line-strength indices versus for the galaxies analyzed in this Letter (data in Table 1). Right column: line-strength indices versus H index. The dashed line indicates the limit for considering very young galaxies in the blue spectral range. The dotted line shows the limit for H > 1.8 å. Blue circles correspond to galaxies in the field, while red squares represent Fornax cluster galaxies. S0 galaxies of both samples are plotted in green. The two galaxies with the highest H values are marked with filled symbols. The star represents the cD galaxy NGC 1399 (see Lyubenova et al. 2008, for an extensive study of this object in the infrared). Symbol size increases according to the central velocity dispersion of the galaxies.

indices against (left column) and H (right column). We note that the number of galaxies analyzed here is small and, therefore, our results might not necessarily apply to all early-type galaxies. There is a clear correlation between the indices Mgb, C4668, Na i, DCO , and for the Fornax galaxies (already shown by Kuntschner 2000 for the optical indices and S08 for the nearIR features). For the field sample, the indices Mgb, C4668 and Na i are also correlated with , and these relations exhibit a larger scatter than those obtained for the Fornax sample (see a quantitative statistical analysis of all these diagrams in Table 2). We do not find differences in the Mgb measurements between both galaxy samples, while field galaxies exhibit higher C4668,


L202

´ ´ MARMOL-QUERALTO ET AL.
Table 2 Statistical Analysis of the Correlations for Field and Fornax Galaxies Diagram rS Figure Figure Figure Figure 2(a): 2(b): 2(c): 2(d): Mgb­ C4668­ Na i­ DCO ­ 0.7636 0.6455 0.6091 -0.0545a 0.9818 0.5182 0.5727 Low Density 0.0031 0.0160 0.0233 0.4367 4E-08 0.0512 0.0328 sr 0.6685 1.0440 1.4256 2.7721 0.0531 0.8005 1.0957 rS 0.5636 0.9273 0.9515 0.5757 0.9879 0.5394 0.6000 Fornax 0.0449 5E-05 1E-05 0.0408 5E-08 0.0538 0.0333 sr 0.4764 0.3287 0.2165 1.5668 0.0787 1.6243 1.9062

Vol. 705

Figure 3(a): Na i­C4668 Figure 3(b): DCO ­C4668 Figure 3(c): DCO ­Na i

Notes. For each diagram, indicated in the first column of this table, the Spearman rank correlation coefficient rS and the associated significance level are given, separately, for the field and Fornax subsamples. In all the cases, the two galaxies with higher H values (filled symbols in the diagrams) have been excluded. With the aim of comparing the scatter observed in each galaxy subsample the residual standard deviation sr has also been determined as the mean quadratic distance from a linear fit to the data points. Since the data have uncertainties in both axes, the fits have been computed using an ordinary least squares (OLS) bisector method (Isobe et al. 1990). The data ranges were previously renormalized for the proper measure of the perpendicular distances from the data to the fitted straight lines. This renormalization prevents the different weighting introduced by the absolute values of the data. Note that, despite the relations between the indices are evident from the figures, some correlations are not significant due to size of the samples. a If the field galaxy with lowest , NGC 3605, is also excluded, we obtain r = -0.4061 ( = 0.1221). S Regarding its position in the Mgb­ and C4668- diagrams, this galaxy could belong to the new family of high-metallicity low-luminosity galaxies found by Sansom & Northeast (2008).

Na i, and DCO indices than Fornax galaxies at < 200 km s-1 , with the largest differences for the CO feature. This behavior indicates that there are differences in the star formation history and the metal enrichment between both galaxy samples. As we will discuss below, we are likely seeing the effect of more metal rich intermediate-age stellar populations in these field galaxies, which are not present in most of their Fornax counterparts (S08). For galaxies with higher , both samples seem to follow the same trend. In the right column of Figure 2 we compare different indices with H . We do not find any correlation. Two of the galaxies-- marked in all the plots with filled symbols--show very high H indices. These galaxies deviate from all the relations defined by the indices and other parameters. In addition, there are several galaxies with 1.8 å < H < 2.8 å, while the bulk of the galaxies have lower H indices. S08 already showed the presence of a small amount of intermediate-age population in the three Fornax galaxies with H > 1.8 å. Since we find differences in C4668, Na i, and DCO between galaxies in low-density environments and in the Fornax cluster, we explore possible dependences among these indices in Figure 3. The most interesting result is the extremely tight correlation between Na i and C4668 indices in Figure 3(a). This is remarkable given the fact that the indices have been measured on completely independent spectra processed by different authors in distinct spectral ranges. Interestingly, the two galaxies with highest H separate from the global trend. Finally, there are also correlations between DCO and the indices C4668 and Na i (Figures 3(b) and 3(c); see Table 2). 4. DISCUSSION The main two results of this work are (1) the excellent correlation between the Na i and C4668 indices, and (2) the higher DCO and Na i values in the low-sigma field galaxies compared with their Fornax counterparts. How can we interpret these results in terms of physical parameters? To answer this

Figure 3. Index­index diagrams for the galaxies in this study. Symbols are explained in Figure 1.

question, we need to take into account the known dependences of the indices to several parameters. We have to warn the reader that this section is rather speculative, because of the lack of stellar population models in this wavelength region. We should also note that a simple discussion in terms of simple stellar populations (SSPs) is inadequate here, since, as we will discuss, the measured indices in Fornax and field galaxies are very likely indicating differences in their star formation history. First we describe the qualitative behavior of the indices. For an old SSP, we deduce from the work by MQ08 that DCO increases as a function of metallicity. For younger stellar populations,


No. 2, 2009

INTERMEDIATE-AGE STELLAR POPULATIONS IN EARLY-TYPE GALAXIES

L203

DCO increases due to the presence of AGB stars. The Na i index increases with metallicity for old stellar populations (see S08). In S08 it was also shown that Na i increases when a young population is present in the galaxy (see Figure 16 in that paper). Mgb and C4668, like the rest of the non-Balmer indices in the Lick system, are weaker in younger stellar populations (e.g., Vazdekis et al. 2003; Bruzual & Charlot 2003; Thomas et al. 2003; Schiavon 2007). These indices are also sensitive to metallicity, with higher values for more metal-rich populations. How to explain the behavior of the line indices shown in Figure 2? Following previous studies analyzing galaxies in different environments (e.g., SB03 Trager et al. 2000; Kuntschner ´ ´ et al. 2002; Thomas et al. 2005;Sanchez-Blazquez et al. 2006a, 2006b; Collobert et al. 2006), we start with the assumption that the differences between the field and Fornax samples could be due to distinct star formation histories. As Kuntschner (2000) showed for the Fornax cluster, elliptical galaxies and lenticulars of high form an almost coeval family of galaxies, and they exhibit a wide range in metallicity which explains the good correlation of all the indices with . On the other hand, previous works have suggested an extended star formation history in field galaxies (e.g., SB03 Collobert et al. 2006; Toloba et al. 2009). If this is the case, we would expect two effects. (1) The chemical enrichment of the interstellar medium produced by previous generations of stars. As a result, the new generation of stars will be more metal rich than the previous ones. (2) The presence of AGB stars from the intermediate-age population of recent star formation episodes. The higher DCO values for field galaxies with < 200 km s-1 can be explained by a late stage star formation in these galaxies, which is reflected in a more metal rich population and the presence of AGB stars. Since this index has a positive response to both effects, the differences between field and Fornax galaxies are more prominent here than in other indices. In addition, the more metal rich population in field galaxies would be responsible for the high values of Na i and C4668 in these objects (see Figures 2(b) and (c) and Figures 3(a)­(c)), since both indices are very sensitive to metallicity. Actually, the excellent correlation between Na i and C4668 shown in Figure 3(a) might be indicating that the effect of an intermediate-age population in these two indices is less important, and, therefore, they are both excellent metallicity indicators for old and intermediate ages. However, the two galaxies with the highest H separate from the global trend, as expected due to the sensitivity of these indices to the presence of very young and more metal rich stars, as explained above. The larger scatter in DCO versus C4668 diagram (Figure 3(b)) can be explained by considering that DCO is sensitive to the presence of AGB stars, contrary to the optical C4668 index. Last, both galaxy samples exhibit a similar Mgb index. This is explained by the age­metallicy degeneracy described in previous works (e.g., Worthey et al. 1992; Pedraz et al. 1998; JÜrgensen et al. 1999; Kuntschner et al. 2001). 5. CONCLUSIONS This work has shown the crucial role that the near-IR linestrength indices can play in the understanding of the star formation histories of early-type galaxies. Although interpreting these near-IR indices is difficult without models and the analyzed galaxy samples are still small, we have demonstrated that the measurement of line-strength indices in the near-IR is a useful tool to constrain the star formation histories of these

galaxies. In particular, the Na i index has been shown to be a metallicity indicator as good as C4668 in the optical range, while the DCO index can be used as a tracer of intermediate-age stellar populations. The differences in C4668, Na i, and DCO , when studying low-mass galaxies in Fornax cluster and in lowdensity environments, can be interpreted as field galaxies having undergone later stellar formation episodes than Fornax galaxies. Once synthesis stellar population models in the near-IR range are available, more detailed and quantitative predictions will be made using this interesting wavelength region. The authors acknowledge the referee, Bahram Mobasher, for his useful comments. E.M.-Q. acknowledges the SMES for a FPI PhD fellowship. This work has been partially supported by the SMES through the research project AYA2007-67752-C0301. N.C. also acknowledges financial support from the research project AYA2006-15698-C02-02. P.S.-B. acknowledges the financial support from a Marie Curie European Reintegration Grant within the 7th European Community Framework Programme. Facilities: VLT:Antu (ISAAC) REFERENCES
Aaronson, M., Cohen, J. G., Mould, J., & Malkan, M. 1978, ApJ, 223, 824 Allard, E. L., Knapen, J. H., Peletier, R. F., & Sarzi, M. 2006, MNRAS, 371, 1087 Bruzual, G., & Charlot, S. 2003, MNRAS, 344, 1000 Cardiel, N. 1999, PhD thesis, Universidad Complutense de Madrid Cesetti, M., et al. 2009, A&A, 497, 41 Collobert, M., Sarzi, M., Davies, R. L., Kuntschner, H., & Colless, M. 2006, MNRAS, 370, 1213 Frogel, J. A., Persson, S. E., Matthews, K., & Aaronson, M. 1978, ApJ, 220, 75 Frogel, J. A., Stephens, A., Ram´ irez, S., & De Poy, D. L. 2001, AJ, 122, 1896 Isobe, T., Feigelson, E. D., Akritas, M. G., & Babu, G. J. 1990, ApJ, 364, 104 James, P. A., & Mobasher, B. 1999, MNRAS, 306, 199 JÜrgensen, I., Franx, M., Hjorth, J., & van Dokkum, P. G. 1999, MNRAS, 308, 833 Kuntschner, H. 2000, MNRAS, 315, 184 Kuntschner, H., Lucey, J. R., Smith, R. J., Hudson, M. J., & Davies, R. L. 2001, MNRAS, 323, 615 Kuntschner, H., Smith, R. J., Colless, M., Davies, R. L., Kaldare, R., & Vazdekis, A. 2002, MNRAS, 337, 172 Lyubenova, M., Kuntschner, H., & Silva, D. R. 2008, A&A, 485, 425 ´ ´ Marmol-Queralto, E., Cardiel, N., Cenarro, A. J., Vazdekis, A., Gorgas, J., ´ ´ Pedraz, S., Peletier, R. F., & Sanchez-Blazquez, P. 2008, A&A, 489, 885 (MQ08) Mobasher, B., & James, P. A. 1996, MNRAS, 280, 895 Mobasher, B., & James, P. A. 2000, MNRAS, 316, 507 ´ Pedraz, S., Gorgas, J., Cardiel, N., & Guzman, R. 1998, Ap&SS, 263, 159 ´ ´ ´ Sanchez-Blazquez, P., Gorgas, J., Cardiel, N., Cenarro, J., & Gonzalez, J. J. 2003, ApJ, 590, L91 (SB03) ´ ´ ´ Sanchez-Blazquez, P., Gorgas, J., Cardiel, N., & Gonzalez, J. J. 2006a, A&A, 457, 787 ´ ´ ´ Sanchez-Blazquez, P., Gorgas, J., Cardiel, N., & Gonzalez, J. J. 2006b, A&A, 457, 809 Sansom, A. E., & Northeast, M. S. 2008, MNRAS, 387, 331 Schiavon, R. P. 2007, ApJS, 171, 146 Silva, D. R., Kuntschner, H., & Lyubenova, M. 2008, ApJ, 674, 194 (S08) Thomas, D., Maraston, C., & Bender, R. 2003, MNRAS, 339, 897 Thomas, D., Maraston, C., Bender, R., & Mendes de Oliveira, C. 2005, ApJ, 621, 673 ´ ´ Toloba, E., Sanchez-Blazquez, P., Gorgas, J., & Gibson, B. K. 2009, ApJ, 691, L95 ´ Trager, S. C., Faber, S. M., Worthey, G., & Gonzalez, J. J. 2000, AJ, 120, 165 Vazdekis, A., Cenarro, A. J., Gorgas, J., Cardiel, N., & Peletier, R. F. 2003, MNRAS, 340, 1317 Wallace, L., & Hinkle, K. 1996, ApJS, 107, 312 Worthey, G., Faber, S. M., & Gonzalez, J. J. 1992, ApJ, 398, 69