a r X i v :a s t r o -p h /9701036v 1 8 J a n 1997
A new chemo-evolutionary population synthesis model for early-type
galaxies.II:Obrvations and results
A.Vazdekis 1(asv@iac.es)R.F.Peletier 2,1(peletier@astro.rug.nl)
J.E.Beckman 1(jeb@iac.es)E.Casuso 1(eca@iac.es)
1
Instituto de Astrofisica de Canarias,E-38200La Laguna,Tenerife,Spain 2
Kapteyn Instituut,Postbus 800,9700AV Groningen,The Netherlands
ABSTRACT
We prent here the results of applying a new chemo-evolutionary stellar population model developed by ourlves in a previous paper (Vazdekis et al.1996)to new high quality obrvational data of the nuclear regions of two reprentative elliptical galaxies and the bulge of the Sombrero galaxy.Here we
fit in detail ∼20absorption lines and 6optical and near-infrared colors following two approaches:fitting a single-age single-metallicity model and fitting our full chemical evolutionary model.We find that all of the iron lines are weaker than the best fitting models predict,indicating that the iron-abundance is anomalous and deficient.We also find that the Ca I index at 4227˚A is much lower than predicted by the models.We can obtain good fits for all the other lines and obrved colors with models of old and metal-rich stellar populations,and can show that the obrved radial gradients are due to metallicity decreasing outward.We find that good fits are obtained both with fully evolutionary models and with single-age single-metallicity models.This is due to the fact that in the evolutionary model more than 80%of stars form within 1.5Gyr after the formation of the galaxies.The fact that slightly better fits are obtained with evolutionary models indicates the galaxies contain a small spread in metallicity.
Subject headings:Chemical Evolution,Elliptical Galaxies,Galaxies:evolution,Galax-ies:formation,Galaxies:abundances,Galaxies:elliptical and lenticular,cd,Galaxies:stellar content,Metallicity,Spectral Energy Distribution,Stellar Evolution,Stellar Spec-troscopy
说明书英文1.INTRODUCTION
The study of the stellar populations and the dis-tribution of metallicities plays an important role in our
understanding of the star formation history of the galaxies.Their stellar populations are expected to be more complex than tho of,for example,globular clusters,which are thought to be compod of a sin-gle stellar population.In fact,Burstein et al.(1984) found differences when comparing colors and spectro-scopic features of globular clusters with galaxies.Key parameters in the interpretation of the obrved col-ors and the line-strengths are the metallicity and the age.The problem is that even in the simplest unre-solved stellar systems their effects are very difficult to parate using only colors(O’Connell1986,Renzini 1986,Buzzoni et al.1992).
Using colors together with absorption lines more accurate conclusions can be drawn.Although ev-ery absorption line strength is dependent on differ-ent kinds of stars,in principle it should be possible to determine average metallicities or ages by care-fully lecting features which more nsitive to the metallicity and others which are more nsitive to the Worthey et al.1992).However the abun-dances of some elements may well evolve differently from tho of α-enhancement),and the conversion of ages and metallicities through models to obrved colors and indices may be not unique due to problems stellar evolution theory.Finally, the large velocity broadening in giant elliptical galax-ies implies that only the strongest lines can be ud to obtain physical information from their spectra.
In the process of understanding the stellar popu-lation of early-type galaxies wefirst developed a new spectrophotometric model,which can be ud to in-terpret obrved colors and absorption lines of galax-ies(Vazdekis et al.1996,hereafter Paper I).The model is bad on the latest improvements in stel-lar evolution theory and on the most recent stellar li-braries.Instead of studying a large sample of galaxies using a few lines indices,as has been done Worthey et al.1992,Gonzalez1993),we preferred to obtain high quality obrvations of three repren-tative early-type galaxies(two giant ellipticals and the bulge of the Sombrero galaxy),but in many col-ors and absorption lines,and to make very detailed fits to each index,to understand better global ages and metallicities,and also to follow the abundances of individual elements.Such analysis now is possible,since we could calibrate our obrvations using the large sample of stars from the extended Lick-system (Worthey et al.(1994)hereafter WFGB).
In this paper we have applied our spectrophoto-metric population synthesis model following both the single-age single-metallicity and the chemical evolu-tionary approaches.We address here the problem of whether the conclusions we obtain depend on the stel-lar population synthesis method we u.In the end wefind that the u of many indices does yield in-teresting information,and we show that we can learn more than by using only a few indices,as has been done in the past.At the same time we study the stellar population gradients in the three galaxies.
This paper is organized as follows:in Section2we explain our obrvations and the method we u to derive the line-strengths.In Section3wefit our pop-ulation synthesis model and discuss the results ob-tained byfitting the data.Finally in Section4we prent our conclusions.
2.Obrvations and data reduction
2.1.Obrvations
Long-slit spectra of three well know early-type galaxies were obtained with the ISIS spectrograph on the4.2m WHT in March,1995,at the Obrvatorio del Roque de los Muchachos,La Palma.The spectra were taken using both arms of the instrument,with a large format TEK windowed to1124×600pixels, each with a size of24µm CCD chip.In the blue we ud a grating of600lines/mm giving a sampling of 0.79˚A pix−1while in the red we ud a300lines/mm grating giving a sampling of1.46˚A pix−1.Our spectra were taken in the range3700−6300˚A.This configu-ration allowed us to cover almost the whole t of the absorption features contained in WFGB in addition to some UV features as defined in Pickles(1985).In fact,in the blue arm we covered the range3700to 4500˚A,while in the red arm we covered the range 4800to6300˚A.We could not obrve the Ca4455, Fe4531and Fe4668features becau they fall in the crossover region of the dichroic.The measured r
eso-lution was∼3.4˚A in the blue and∼6.5˚A in the red spectra.We also lost the TiO2since this index falls at the limit of the range covered.A t of stars of the sample of WFGB was also obrved to calibrate our line-strength measurements.We positioned the slit on the major axis for NGC4472,at123◦for NGC3379 (the major axis is at∼70◦,e Peletier et al.1990a)
and on the minor axis for the Sombrero galaxy.The exposure times were1800s for both frames.
2.2.Data reduction
All data reduction was done with the IRAF soft-ware package.Thefirst step was subtracting the bias value calculated from the unilluminated por-tion of each frame.After this weflat-fielded using Tungsten lamp exposures.Next,the spectra were wavelength calibrated using CuAr+CuNe calibration lamp exposures.The obtained pixel scale was57.535
kms−1pix−1for the blue and79.926kms−1pix−1for the red spectra.The following step was the sky sub-traction,for which the outer parts of each galaxy frame(chon at approximately2′from the center of the galaxy)were averaged to produce a mean sky spectrum,which then was subtracted from each frame.The last step is the elimination of pixels af-fected by cosmic rays in each frame.
2.3.The line-strength measurements
We need a well-defined way to assign values to the strengths of features at different radii in the galaxy. For this purpo we ud the expanded Lick-system (WFGB).Here some indices(the atomic features)are defined as equivalent widths and some(the molecular bands)as ratios of line-depth to continuum in mag-nitudes and we have maintained the definitions.To measure the line-strength indices along the major axis we had to de-redshift each feature and the con-tinuum on each side of it using the recession veloc-ity corresponding to each spectrum.To calculate the rotation curve the spectra at each radius were cross-correlated with the obrved spectrum of a stellar ve-locity standard star which looks most like the galaxy (with spectral type K III,e Section2.3.1).This method is described in Bottema(1988),following the paper of Tonry&Davis(1979).Then we calculated the indices by co-adding a sufficient number of spectra in the spatial direction,so that a satisfactory contin-uum level was reached across the whole wavelength range.More details about the indices can be found in WFGB and Burstein et al.(1984).
2.3.1.The conversion to the expanded Lick-system
aftereffectSince we want to compare our data with the spec-trophotometric model we developed in Paper I,which is bad on the Lick-system,we need to transform our results to that system.Among the problems encoun-
Correction factors
Vel. Lick UV CN
1.016-tail是什么意思英语
FeI+CN
1.036-0.028 CN2
1.334-0.35
featured是什么意思G-band
1.074-
2.95
below是什么意思Hβ
1.1320.37
Mg1
1.006-0.053
江门翻译Mg b
1.1340.22
Fe5335
1.234-0.05
Fe5709
1.2420.0 NaD
1.0670.006
Errors
Poisson Rot.Curve Vel.Disp.Conv.Lick UV CN
经典影片
0.440.060.02-
FeI+CN
0.0050.0010.0010.017 CN2
0.110.060.040.23
G-band
0.250.090.00.46
等等日语
Hβ
0.360.300.020.33 Mg1
0.0020.0030.0010.006 Mg b
0.170.080.020.20
Fe5335
0.150.030.010.14
Fe5709
0.090.150.00.15 NaD
0.0030.0010.0010.005
Fig. 2.—The indices of NGC3379obtained in the UV region.Fig.3.—The visible indices along the major axis of NGC4472.画材
北大青鸟培训
Fig.4.—The UV indices of NGC4472.Fig.5.—The visible indices along the minor axis of Sombrero galaxy.
