Analysis of the structure and central properties of a sample of massive bulgeless galaxies

Bruno Ribeiro Analysis of the structure and central properties of a sample of massive bulgeless galaxies Departamento de F´ısica e Astronomia Faculdade de Ciˆencias da Universidade do Porto Setembro de 2012 Bruno Ribeiro Analysis of the structure and central properties of a sample of massive bulgeless galaxies Tese submetida `a Faculdade de Ciˆencias da Universidade do Porto para obten¸c˜ao do grau de Mestre em Astronomia Departamento de F´ısica e Astronomia Faculdade de Ciˆencias da Universidade do Porto Setembro de 2012 ”[S]ince every piece of matter in the Universe is in some way affected by every other piece of matter in the Universe, it is in theory possible to extrapolate the whole of creation - every sun, every planet, their orbits, their composition and their economic and social history from, say, one small piece of fairy cake.” Douglas Adams, The Restaurant at the End of the Universe ii To Ana, The particle to my antiparticle, The sun to my planetary system, The dark matter to my galaxy, The universe to my life. iii Acknowledgments First of all I would like to give a special thanks to my advisers Catarina Lobo and Sonia Ant´on for providing me guidance and knowledge without which this thesis would not be possible. A grateful thanks to Jean Michel Gomes who collaborated with me in this thesis by providing stellar synthesis population results from STARLIGHT 1for the samples, and also to Polychronis Papaderos for helpful comments and discussion. An acknowledgement to the nancial support from project PTDC/CTE-AST/105287/2008 from FCT. I would also like to acknowledge Chien Peng for making the GALFIT code publicly available and for the very comprehensive explanation on how it works on its website. A special acknowledgment to all people who are part of the Sloan Digital Sky Survey 2(SDSS) project. Last, but not least, I thank Ana for all the support provided during this last year and for some insightful discussions on this matter. 1The STARLIGHT project is supported by the Brazilian agencies CNPq, CAPES and FAPESP and by the France-Brazil CAPES/Cofecub program 2Funding for the SDSS and SDSS-II has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, the U.S. Department of Energy, the National Aeronautics and Space Administration, the Japanese Monbukagakusho, the Max Planck Society, and the Higher Education Funding Council for England. The SDSS Web Site is http://www.sdss.org/. iv Preface In the beginning there were us, the sun and the moon. Shortly after, the planets joined the big picture, and stars, well, stars were just like light bulbs in the walls of the universe. We evolved, and along with us, so did technology Soon, new planets were discovered, the nature of stars was revealed, we were part of something bigger than ever thought possible. Years and years had gone by and we had a universe of our own, our galaxy. Like two fried eggs, sunny-side up, put together, back to back, with a disk , the glair, a bulge, the yolk, and the tiny drops of hot oil bouncing moving around like the globular star clusters of the halo. However, some spots, not like stars, in that perfect and heavenly world puzzled astronomers. Could there be other worlds like our own spread in a larger universe? In the early 20th century, we embraced that idea. There were other worlds outside our own, thousands and thousands of them, and as different from each other as humans are. Ones were giant balls of old stars without traces of dust and gas, others more like our own galaxy and with young and newly born stars, some with beautiful spiral patterns designed by their gas and dust harboring stellar maternity wards, and there are even some which appeared as a scrambled mix of dust, gas and stars, places of star birth. As humans, different galaxies belong to different places. The big old ones live together in big groups accompanied by small ones most like them, and the younger were roaming free through the fields of the universe or remained in the outskirts of the elder groups. The universe seems beautiful, an organized and wonderful place to live in. But we have yet to understand all of its wonders and mechanisms. v Abstract The aim of this thesis is to study a sample of red massive bulgeless galaxies, selected from the SDSS DR7 based on an automated algorithm that performed one-dimensional analysis of the galaxy light profiles. From an initial sample of 77 bulgeless candidates we found 38 bulgeless galaxies and 29 galaxies with a pseudo-bulge regarding their large-scale structure using two-dimensional modeling techniques. An additional sample of 20 bulge-dominated candidates was selected to serve as a control sample and all of them were confirmed as bulgy galaxies after the two-dimensional analysis. We found that the disks of pseudo-bulge galaxies have larger effective radius than the bulgeless galaxies. Additionally, using the SDSS optical spectra, we assess some physical properties of the central regions of these galaxies (central 3”). We find that the bulgy galaxies show different properties (stellar mass, metallicity, mean stellar age, dust content) from bulgeless and pseudo-bulge galaxies and that the distinction between the last two classes of galaxies is only possible when considering central stellar mass and central velocity dispersions. vi Contents Preface iv Abstract v List of Figures ix List of Tables x 1 Introduction 1 2 Description of the sample 4 2.1 The bulgeless nature of galaxies . . . . . . . . . . . . . . . . . . . . . . 7 2.2 ControlSample ............................... 9 3 Structure analysis of the sample galaxies 14 3.1 Detailed Decomposition of Galaxy Images using GALFIT . . . . . . . . 15 3.1.1 GALFITfunctions ......................... 17 3.1.2 Originalimages........................... 18 3.1.3 Testing for variations of the input point-spread function (PSF) . 18 3.1.4 Models used in the image decomposition . . . . . . . . . . . . . 19 3.1.5 Input files and parameters . . . . . . . . . . . . . . . . . . . . . 22 3.1.6 GALFIT based exclusion . . . . . . . . . . . . . . . . . . . . . . 24 vii 3.2 Construction of surface brightness profiles . . . . . . . . . . . . . . . . 25 3.2.1 STSDAS ellipse routine ...................... 26 3.3 StructureResults.............................. 27 3.4 Comparison between galaxies with and without a significant bulge component.................................... 33 4 Stellar content of central regions 43 4.1 STARLIGHT ................................ 44 4.2 Inferred physical properties . . . . . . . . . . . . . . . . . . . . . . . . 45 4.3 Results.................................... 47 5 Summary, Discussion & Conclusions 53 5.1 Discussion & Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 53 5.2 Futurework................................. 56 A Excluded objects and bulgy sample 57 B GALFIT input file 61 C GALFIT residuals and surface brightness profiles 63 References 86 viii Chapter 2 Description of the sample The main focus of this study is to establish a detailed structural description of a set of galaxies in order to confirm them clearly as a class of red bulgeless candidates. Further analysis are also performed to characterize other properties of these galaxies (stellar populations, star-formation history (SFH), metallicities) to gain insight on the formation of pseudo-bulges.The initial sample from which this study was carried out is issued from the New-York University Value Added Galaxy Catalog 1(NYUVAGC, [8]) and compiled in [15]. The sample selection criteria is summarized in table 2.1 and justified below. Throughout this study, all calculations involving cosmological parameters used H0= 71kms−1Mpc−1,Ωm= 0.27 and ΩΛ= 0.73 (WMAP7, [44]). The lower redshift cut (0.02 < z < 0.06) applied serves as a filter to not consider galaxies with relevant contamination from their peculiar velocities in the determination of the redshift and to exclude possible sources with largely extended morphology from which the recovered SDSS spectra would correspond to the very central part of the galaxy (remembering that the SDSS fiber encloses 3” in the sky plane [62]). The higher limit in redshift was chosen so that the resulting galaxies have reasonable resolution which is very important since we are interested in modeling its morphology, and structural features tend to become unrecognizable in the images as we move towards higher redshifts. Considering the SDSS plate scale of 0.396”/pixel [62] at z= 0.06 one side of a pixel subtends ∼444pc which is smaller than the typical size of the bulge of a galaxy and so we are, in principle, capable of clearly classifying these galaxies as bulgeless candidates. The stellar mass cut was primarily defined to favor the selection of galaxies whose star 1http://sdss.physics.nyu.edu/vagc/ 4 CHAPTER 2. DESCRIPTION OF THE SAMPLE 5 formation might be affected by a given mechanism, AGN feedback, for the purpose of the study of Coelho (2010) [15]. As for the assessment of the bulge’s significance, one of the commonly used ways to distinguish between bulge dominated galaxies and disk dominated galaxies is to fit a S´ersic profile [55] Σ(r) = Σeexp[−κ(r/re)1/n −1] (2.1) where the S´ersic index ndescribes the shape of the light profile, reis the effective radius of the profile, Σeis the surface brightness at radius reand κis a parameter coupled to n[14] such that half of the total flux is enclosed within re(see section 3.1.1). An index of n= 1 corresponds to a typical pure disk galaxy, whereas n= 4 corresponds to the de Vaucouleurs profile associated to elliptical galaxies. The values of navailable in the NYU-VAG catalog were obtained through the fitting of equation 2.1 to azimuthally average radial profile convolved with the estimated seeing for each galaxy [8]. Following Bell (2008) [6], only galaxies with n < 1.5 were selected. This is a commonly used frontier value to select bulgeless galaxies. The limits on color index g−rand on the inclination parameter qam are used to limit our sample to galaxies with little star-formation activity (again, for the original study purpose of the study of Coelho (2010) [15]) . Since younger stellar populations emit strongly on the bluer part of the optical spectra and older stellar populations emit predominantly towards higher wavelengths , the color index serves as an indicator of the relative age of the stellar populations (see Figure 2.1). Following [6], red galaxies are defined as having g−r > 0.57 + 0.0575 log10(M∗/108M). As we are interested in galaxies with low star-formation rates, one needs to distinguish between red galaxies dominated by old stellar populations and dust-obscured galaxies. It is well established that dust is mainly concentrated in the disks of the galaxies [11, Chapter 25]. This means that edge-on disk galaxies have a higher column density of dust that obscures the light of the younger stellar populations leading to an apparently red galaxy. So, setting a limit on the inclination of the galaxy, minimizes the inclusion of dust obscured objects in the sample. This is often made by setting a lower limit on the axis ratio (typically b/a > 0.5 ) for selecting galaxies. This usually works because the greater the inclination of the galaxy typically leads to a decrease in this ratio. However, the approach used in [15], uses the inclination parameter, qam instead, CHAPTER 2. DESCRIPTION OF THE SAMPLE 6 Figure 2.1: Sensitivity functions of the ugriz SDSS filter system. The blue line represents the optical spectra of a typical blue galaxy and the red line the optical spectra of a typical red galaxy. Both spectra were suitably normalized for viewing purposes. [29]. which might be a more robust indicator for the inclination qam =1−E 1 + E1/2 , with E =qm2 1+m2 2(2.2) where m1and m2are the second order adaptive moments of the galaxy’s light profile [60] available in the NYU-VAGC catalog. Table 2.1: Sample selection criteria. (z) Redshift, (M∗) Stellar mass, (n) r-band 1D S´ersic index, (g−r) color index, (qam) inclination parameter Selection limits 0.02 < z < 0.06 M∗>1010M n < 1.5 g−r > 0.57 + 0.0575 log10(M∗/108M) qam >0.5 CHAPTER 2. DESCRIPTION OF THE SAMPLE 7 2.1 The bulgeless nature of galaxies Some of these parameters are too simplistic or occasionally produce wrong selections when applied automatically to large datasets. My work concentrates on the structure of a small set of galaxies: it is a detailed and dedicated analysis to provide confirmation of the bulgeless nature of these objects. Testing whether a galaxy is bulgeless or not is a delicate procedure. So in order to be certain about the structural parameters of the galaxies we wish to study, we need to be cautious about the objects we want to analyze. Bearing this in mind I started with the initial list of 113 objects that resulted from applying the previous criteria (see Table 2.1)and retained only those that passed a thorough visual inspection. The main reasons for the exclusion of an object are: •Presence of dust lanes; •Highly disturbed morphology; •Overlapping of bright objects; •Being the brightest cluster galaxy (BCG) with a bulgy shape; •Presented obviously mis-computed colors. The presence of dust lanes poses a challenge to 2-dimensional modeling of the galaxy since we have to take into account the decrease in brightness that affects the galaxy’s light profile. Despite being possible to adjust a truncated S´ersic model we chose not to do it because it can produce unrealistic values of the S´ersic index since we do not have any central information on the true brightness of the stellar component of the galaxy. This is a conservative approach but we are focused on determining beyond any doubt the bulgeless character of these galaxies and the significant presence of dust may bias the model to unrealistic underlying profiles. Moreover the presence of dust shades doubt on the hypothesis that the red colors of the galaxy is mainly due to it being dominated by old stellar populations. Disturbed morphology (asymmetries, tidal tails, distorted shapes) often indicates a recent or ongoing merger process which is known to likely induce a burst in star formation in gas-rich galaxies [56, Chapter 7]. Since we are interested in galaxies with low star-formation activity we chose not to include these objects in the sample. Also, the model decomposition would involve adding Fourier modes to the underlying models [50] which may affect the distinction between bulgeless and bulgy galaxies. CHAPTER 2. DESCRIPTION OF THE SAMPLE 8 In the case of galaxies SDSS J083055.47+092838.0, SDSS J083404.99+434150.9 and SDSS J110509.38+380408.0 (see figure A.1) there is a bright object overlapping the galaxy’s image (saturated stars in the two last ones) and due to impossible modeling of the separate components we chose to exclude these three galaxies as well. NASA Extragalactic Database 2(NED) classifies galaxies SDSS J004150.47-091811.2 and SDSS J122306.66+103716.4 as BCGs. These were erroneously selected due to having an extended stellar envelope that the automatic algorithm confused it with an extended disk. The list of excluded objects and its exclusion motive are summarized in table 2.2. The color images (gri composite bands) are displayed on Figure A.1. This conservative approach led to a sample of 77 galaxies on which this study will be based. The resulting sample may be consulted in Figure 2.2. Table 2.2: List of excluded objects Name Exclusion Motive SDSS J004150.47-091811.2 BCG SDSS J075816.66+271029.5 Dust lane SDSS J082205.75+562534.4 Dust lane SDSS J083055.47+092838.0 Overlapping brigh object SDSS J083404.99+434150.7 Overlapping brigh object SDSS J084105.25+385439.3 Dust lane and Disturbed Morphology SDSS J084958.78+381203.2 Dust lane SDSS J091322.82+225156.9 Large magnitude errors SDSS J091530.43+543129.7 Dust lane SDSS J100204.32+505437.3 Dust lane SDSS J102154.20+135356.4 Disturbed Morphology SDSS J102238.05+231015.5 Dust lane SDSS J102733.32+102018.9 Dust lane SDSS J105804.22+170836.8 Dust lane SDSS J110509.38+380408.0 Overlapping brigh object SDSS J111718.56+293610.6 Dust lane SDSS J112724.73+273714.2 Disturbed Morphology SDSS J113732.29+092002.2 Dust lane SDSS J114517.35+271634.4 Dust lane 2http://ned.ipac.caltech.edu/ CHAPTER 2. DESCRIPTION OF THE SAMPLE 9 Table 2.2 (continued) Name Exclusion Motive SDSS J120559.45+022953.6 Dust lane SDSS J120800.34+231307.5 Dust lane SDSS J120801.37+325622.0 Large magnitude errors and Disturbed Morphology SDSS J121611.87+592315.2 Dust lane SDSS J121748.54+463454.9 Dust lane & Merger SDSS J122306.66+103716.4 BCG SDSS J123226.76+444339.7 Dust lane SDSS J125227.71+600400.6 Disturbed Morphology SDSS J131148.02+305821.3 Dust lane SDSS J134139.23+554014.0 Disturbed Morphology SDSS J135600.11+173041.7 Dust lane SDSS J143351.99+272042.8 Dust lane SDSS J160753.53+101609.7 Dust lane SDSS J165251.83+360541.2 Dust lane SDSS J170024.77+382115.5 Disturbed Morphology SDSS J223002.77-001652.6 Disturbed Morphology SDSS J230211.71+142829.0 Dust lane 2.2 Control Sample For comparison and to test the performance of the structure-determination algorithm and other methods used in this work, besides selecting the main sample for this study consisting of bulgeless galaxies as assessed by the 1D profile fitting, an additional set of bulge-dominated galaxies was assembled. The bulgy galaxies obey the same selection criteria as the bulgeless sample in what regards the redshift interval, the stellar mass and the g−rcolor index. No selection was performed based on the inclination parameter because we do not expect dust to be very abundant in this class of objects [56, Chapter 6] so a selection on their inclination is not relevant for our purposes. Only galaxies with a S´ersic index of 4 <n<5.5 were selected. This interval of S´ersic indexes was defined as such so that these galaxies had preferably no disk component to assess the differences in general properties that may arise. These were all elliptical galaxies. From all the galaxies selected in this way, twenty were CHAPTER 2. DESCRIPTION OF THE SAMPLE 10 Figure 2.2: Color (gri bands) images of the 77 galaxies in the final sample. CHAPTER 2. DESCRIPTION OF THE SAMPLE 11 Color (gri bands) images of the 77 galaxies in the final sample. (continued) CHAPTER 2. DESCRIPTION OF THE SAMPLE 12 Color (gri bands) images of the 77 galaxies in the final sample. (continued) CHAPTER 2. DESCRIPTION OF THE SAMPLE 13 Color (gri bands) images of the 77 galaxies in the final sample. (continued) chosen by visual inspection to guarantee the goodness of the selection criteria (see Figure A.2). This control sample serves the purpose of investigating the recovered values by the two-dimensional modeling, to test if galaxies with a prominent bulge can be misclassified as bulgeless by this procedure and to compare with the results of the bulgeless sample so that we can test if they have different general properties. CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 20 Table 3.1: Summary of the results for the PSF variation test. The presence of two S´ersic indexes means the model has two components. The mean value refers to the mean of the three output values for each input PSF. Name Parameter Mean Value Standard deviation SDSS J154408.74+012541.8 FWHMa2.54 0.01 n1.17 0.05 FWHMa3.20 0.13 SDSS J020251.99-080136.1 n10.91 0.06 n21.00 0.02 FWHMa2.56 0.10 SDSS J221917.33-011113.7 n10.84 0.00b n20.50 0.04 aFull Width at Half Maximum (in pixels) retrieved from IRAF routine imexamine for the PSF. bAll three cases have the same index. Figure 3.1: SDSS r-band images for the three galaxies used for the PSF test. From left to right: SDSS J154408.74+012541.8, SDSS J020251.99-080136.1 and SDSS J221917.33-011113.7. The offset in the middle image has been applied in order to mask out a saturated star. CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 21 Figure 3.2: GALFIT output residuals for the modeling of SDSS J154408.74+012541.8 using three different input PSF images. The distance of the PSF to the galaxy increases from left to right. Figure 3.3: Same as figure 3.2 for SDSS J020251.99-080136.1 Figure 3.4: Same as figure 3.2 for SDSS J221917.33-011113.7 CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 22 work of Gadotti (2009) and others ( [30], [5]), where a similar method was applied, the selection cut on the index was a little higher (n≤2). Since the value of nvaries continuously from a Gaussian profile (n= 0.5) to an exponential disk (n= 1) and to a classical bulge (n= 4) it is not easy to set a limit to separate bulgy galaxies from those with no bulges or with pseudo-bulges. But n < 1.5 should be conservatively safe. In this work I designed two strategies to study the global structure of these galaxies and tackle the problem of their morphological classification. Firstly, a n-free S´ersic profile was fitted to all galaxies and a second n-free S´ersic component was added whenever the residuals indicated some remaining coherent structure. Secondly, and in alternative, a fixed S´ersic profile with n= 1 (exponential disk) was fitted to all galaxies and a second n-free S´ersic profile was added only when necessary. Despite knowing that no two galaxies are alike and that restricting the value of one of the S´ersic indices leads to a bias in the distribution of the free nof the second component,, there is degeneracy of the profile due to variations in the other parameters, namely the effective radius, re, which may cause two profiles with different S´ersic indices to adjust equally well the same galaxy. I have tested and I am confident that the second strategy is more robust against possible variations in the parameters and leads to more physically meaningful results. Thus, the classification of galaxies obtained in this way is more reliable and, therefore, the second model shall be preferentially adopted in this work. The final classification scheme for these galaxies divides them in three main groups: •Bulgeless Galaxies - well adjusted by a single exponential disk; •Pseudo-bulge Galaxies - galaxies which require a second component with n < 1.5 •Bulgy galaxies - galaxies which are best modeled with a n-free S´ersic profile with n > 2 3.1.5 Input files and parameters In order to model a galaxy, GALFIT needs a set of input images and values to produce reliable results. The input image is, by default, a section of the field centered on the galaxy to fit (whenever not possible - because some galaxies are close to the CCD edge or have CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 23 Figure 3.5: Mask applied to SDSS J161159.399+300251.7 image on the left and SDSS r-band image on the right. The white pixels represent the bad pixels which GALFIT will not take into account. saturated objects nearby - an offset was applied provided that all the galaxy’s emission was still contained in the region and any problematic object excluded). The limits of the region were defined so that it can have a significant portion of sky in it while minimizing the number of objects present there. Whenever a saturated star nearby could not be excluded from the region to fit, a mask file preventing GALFIT to account for the saturated pixels was provided. The mask is a polygon region surrounding a problematic object defined using ds9 [35] regions tools and then converted to a bad pixel list using the algorithms provided in the help pages of GALFIT 5. One can see an actual mask used for galaxy SDSS J161159.399+300251.7 in figure 3.5. The PSF input image is an isolated, bright and non-saturated star from the field as close to the galaxy as possible, from which the sky value present in the image header was subtracted. The magnitude photometric zero point and the sky level were retrieved from the image header of each field. The plate scale of the CCD can be found in [62]. Despite being somewhat insensible to the initial set of parameters for the S´ersic function in the case of simple modeling by one component, one must not provide unrealistic values since that compromises convergence. In that sense, rough estimates of xc,yc,re,b/a and θPA were retrieved with the help of ds9 tools. The initial value for nwas taken from the NYU-VAGC and mtot was taken from the SDSS navigate tool information 6. 5http://users.obs.carnegiescience.edu/peng/work/galfit/MASKING.html 6http://cas.sdss.org/dr7/en/tools/chart/navi.asp CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 24 In the case of adding a second component, and because the first component usually fits the extended emission of the galaxy, the initial parameters were set as follows: xc, yC,θPA and mtot were the same; b/a was set to a higher value close to 1 to reflect the possible existence of a spheroidal component; nwas set to a higher value to reflect the steeper central profile characteristic of a spheroidal component and rewas set to a lower value as we expect the spheroidal component of the galaxy to be more concentrated than the disk. In the particular case when a first component was fixed to be an exponential disk (n= 1), the second component was set to have an initial value of n= 4, that of a classical bulge. Also, since in some cases there are nearby galaxies and stars that are included in the region to fit, I enabled GALFIT to model these objects too (as indicated in section 3.1.1), to minimize contamination from light of other objects other than the galaxy we are studying. An example of an input file is given in Appendix B. 3.1.6 GALFIT based exclusion The GALFIT results enable us to re-assess more confidently the structure of the galaxies in the departure sample of 77 objects. Based on those results, we further narrowed the sample as justified below. Three objects, SDSS J154408.74+012541.8, SDSS J230751.49+142333.5 and SDSS J074600.04+214323.2, were excluded since they are very faint sources. Consistently, their SDSS spectra are very noisy. Three other objects, SDSS J162534.52+285129.0, SDSS J110810.87+385717.0 and SDSS J112534.58+523247.0, were excluded because they present dust lane features in the r-band images that are also evident in GALFIT residuals, as can be seen in Figure 3.6. Finally, four objects,SDSS J113303.66+354656.7, SDSS J091703.88+264552.3, SDSS J160217.56+162156.6 and SDSS J081931.52+183325.2, were excluded due to the high values of the S´ersic index retrieved from GALFIT modeling using a single n-free S´ersic model (n > 4 for all of them) as the best fit. These were also confirmed when trying a combination of disk+S´ersic models where the second component turned out to dominate the fit (exponential disk would become too faint in the presence of the second n-free S´ersic component) and presented a S´ersic index always greater than 3. CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 25 Figure 3.6: GALFIT residuals for the three galaxies presenting dust lane features. From left to right: SDSS J112534.58+523247.0, SDSS J162534.52+285129.0 and SDSS J110810.87+385717.0. After this re-assessment, 67 galaxies were selected as the bulgeless and pseudo-bulge candidates on which this thesis results are based on. 3.2 Construction of surface brightness profiles As described in the beginning of this chapter, one of the ways to quantify the structure of a galaxy is to adjust a function to its one-dimensional light profile. In this work one-dimensional light profiles of the galaxies were computed, though no fitting was performed, and compared with that of the models to provide a better visual inspection of the profiles. The reason behind not simply using the GALFIT output parameters to describe the models is because the models are convolved with a PSF for comparison with the original galaxy image. If one wishes to reproduce the same results in a one-dimensional fashion using directly the GALFIT output parameters, the S´ersic function must be convolved with the correspondent one dimensional PSF model, or a deconvolution of the galaxy’s image has to be performed. Since deconvolution of the galaxy’s image depends on high S/N and one-dimensional convolution is mathematically different than the two-dimension convolution performed by GALFIT I chose to extract the one-dimensional profiles directly from the output image models provided with the GALFIT results. This was made using the ellipse routine designed for the Image Reduction and Analysis CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 26 Facility 7(IRAF) and part of the Space Telescope Science Data Analysis System 8 (STSDAS) package. 3.2.1 STSDAS ellipse routine This routine, described in detail in [34], is used to produce one-dimensional surface brightness profiles from two-dimensional images. I will next describe briefly how the isophote fitting is performed by this task. All my work was performed in interactive mode so visual inspection of the fitting, especially in the outer regions of the galaxy, could be done to control the performance of the algorithm and its results. The fitting starts with a set of four initial parameters: the center of the isophote, given by xcand yc, the ellipticity, defined as = 1 −b/a, and position angle, θ. Then it proceeds to a least-squares minimization of the fitting function (a Fourier transform of I(r)) I(θ) = I0+ 4 X n=1 Ansin(nθ) + Bncos(nθ) (3.5) where I0is the intensity of the isophote, and the parameters A1,A2,B1and B2 are the amplitude of the harmonics whose value is a measure of how much the input parameters of the ellipse (each one relates to a specific parameter) are wrong, i.e. small values of this amplitudes mean that the input parameters are correct and the larger the values the greater the error of the parameter in relation to that which represents the isophote we are considering. Parameters with n= 3,4 give information on how much the isophote deviates from a true ellipse. The correction factors of the ellipse parameters for a specific isophote are computed as follows: ∆xc=−B1 I0 ∆yc=−A1(1−) I0 ∆=−2B2(1−) a0I0 ∆θ=2A2(1−) a0I0[(1−)2−1] (3.6) 7http://iraf.noao.edu/ 8STSDAS is a product of the Space Telescope Science Institute, which is operated by AURA for NASA and available through http://www.stsci.edu/institute/software hardware/stsdas CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 27 where I0is the derivative of the intensity along the major axis evaluated at the semimajor axis length of the isophote being considered, a0. The parameter with the greatest deviation is changed according to equations 3.6 and a new minimization is calculated until convergence is reached. For a good fit the value of I(θ) should be constant for all values of θ. Then it proceeds to the next value of semi-major axis, a0, length by an increment step indicated in the input file. Once the outer regions of the galaxy are reached, i.e. the mean isophotal intensity approaches 0, I reverse the isophote fitting to start calculating the isophotes in the inner regions (with a0< ai where, for most cases, ai= 10pixels). Then the process continues automatically (but still visually inspected) since for the inner regions the intensity values are high enough to perform consistent calculations. To exclude light excess in the outer regions of the profiles due to contamination by nearby bright stars the same mask used in GALFIT modeling was applied in the ellipse routine. There are other output parameters that are not described here because they were not used as I am only interested in obtaining the brightness profiles of the galaxies. The output, I(r), was then converted to represent a surface brightness profile by µ(r) = −2.5log10 I(r) s2texp +mzpt (3.7) Where srepresents the scale factor for conversion between pixel and arcseconds. This procedure was done for sky-subtracted galaxy images and GALFIT output models (for separate components whenever necessary). 3.3 Structure Results The final GALFIT best model for each galaxy was chosen based mainly on two criteria. The first imposing that ∆n/n < 15% (where ∆nis the associated error resulting from GALFIT) and the second based on the appearance of the output surface brightness profiles. The results are summarized in table 3.2. Since one cannot expect that simple S´ersic laws will model every detail of every galaxy, most of them present noticeable residuals. There are cases where residual light is over 10% of its original value. However these pixels refer either to galaxies substructures (mainly spiral arms and sometimes small clumps) and to small central sources whose extent is typically that of the size of the PSF image and therefore refer to an emission that is not adjustable CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 28 Figure 3.7: Surface brightness profile (left) and residual image from GALFIT (right) for galaxy SDSS J003018.19-003008.1. The solid line represents the exponential disk model and the open circles are the galaxy data. The scale of the residual image was set to zscale in ds9 so that the pixels around the median intensity of the image clearly stand out and we can have a better view of the distribution of the residuals. However they are actually quite faint. In this case the differences between the model and the actual profile reflect the spiral pattern of this galaxy. by a S´ersic profile. I further stress that in the brightest central area of the galaxies such residuals are always below the 10% level. Another key aspect to keep in mind when selecting the final model is that galaxies do not naturally fall into the single exponential disk category, i.e. some deviations from this analytical model should be expected and are reflected in the actual results. In other words, few galaxies are matched exactly by an exponential disk. One example is presented in Figure 3.7 where the spiral pattern produces a wobbly disk profile where the zones where the model counts are below the galaxy flux locate the spiral arms and the zones where the model counts are above the galaxy flux pinpoint the regions inbetween arms. Asymmetrical features, ring-like features, inner disks, bars and spiral patterns, which are commonly observed in this sample, will affect the shape of the surface brightness profiles. Thus, the choice between a simple exponential disk model (hereafter model 1) and an exponential disk plus a n-free S´ersic model (hereafter model 2) is not trivial. My choice is based on the visual inspection and physical meaning of the parameters of the two models used (with both of them having ∆n/n < 15%). Whenever the difference between model 1 and 2 did not show significant improvement (mainly because one of the components would become too faint or its effective radius CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 29 Figure 3.8: Surface brightness profiles of models 1 (left) and 2 (right) for SDSS J135857.84+581406.9. In this case the differences between models 1 and 2 are negligible. was greater than that of the exponential disk) model 1 was chosen. There are also the special cases (23 out of 67 galaxies) on which GALFIT could not converge for model 2, meaning that the residuals after exponential disk subtraction have no structure coherent with an extra S´ersic model. Of the other 44 galaxies, 15 show no considerable difference between model 1 and 2 (see Figure 3.8 for an example). Finally, there are two particular cases in which an exponential disk plus a n-free S´ersic profile completely failed to model the galaxy. SDSS J111554.17+204438.4 (see Figure 3.9) is better modeled by the combination of an exponential central component with a faint, low S´ersic index (n= 0.12 ±0.01) extended disk. It still remains on the sample because, despite having no exponential large scale disk component, its central region its far from being a classical bulge and so it falls into the category of a galaxy with a pseudo-bulge. SDSS J170630.27+220003.9 (see Figure 3.10) has a double n-free S´ersic profiles to describe its structure with a bulge component having a value of n= 1.97 ±0.18. Despite being over the imposed n < 1.5 limit of the selection criteria, I chose to keep this galaxy in the sample because its nvalue still remains below 2. I found a galaxy with a truncated profile, SDSS J131659.28+074326.4 (Figure 3.11). The break in the light profile is visible in the GALFIT output residuals just outside the spiral arms where a darker annular region stands out. Other galaxies present in the same field were also modeled by GALFIT and no similar features were found giving indication that this is a real feature of the galaxy light profile and not some artifact CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 36 Figure 3.16: Comparison between the GALFIT based axis ratio and (1) the inclination parameter qam (top panel): (2) the axis ratio from NYU-VAGC (bottom panel). Galaxies modeled with only one component are represented by triangles and the exponential disks of galaxies with two components are represented by open circles. CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 37 Figure 3.17: Normalized histogram of effective radius values for the bulgeless exponential disks (solid line) and for the exponential disks of pseudo-bulge galaxies (dashed line). Figure 3.18: Normalized histogram of the absolute r-band magnitude for the bulgeless exponential disks (solid line) and for the exponential disks of pseudo-bulge galaxies (dashed line). CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 38 Table 3.2: Results of the structural analysis done with GALFIT for the 67 galaxies. (red) Effective radius of the single/disk component. (nd) S´ersic index of the single/disk component. (b/ad) Axis ration of the single /disk component. (reb) Effective radius of the pseudo-bulge component. (nb) S´ersic index of the pseudo-bulge component. (b/ab) Axis ratio of the pseudo-bulge component.(χ2 ν) Reduced χ2value of the fit. Name red[kpc] ndb/adreb[kpc] nbb/abχ2 ν SDSS J003018.19-003008.1 6.32 ±0.06 1.00 0.59 ±0.005 1.039 SDSS J010303.55+132950.3 3.35 ±0.02 1.00 0.58 ±0.005 1.063 SDSS J011500.27+000151.3 5.22 ±0.09 1.00 0.36 ±0.005 0.96 ±0.01 0.86 ±0.02 0.90 ±0.01 1.05 SDSS J011834.14-001341.7 5.90 ±0.05 1.00 0.42 ±0.005 1.38 ±0.01 0.35 ±0.01 0.57 ±0.005 1.16 SDSS J014338.61+133139.6 14.12 ±0.23 1.00 0.14 ±0.005 1.62 ±0.01 0.46 ±0.02 0.96 ±0.01 1.33 SDSS J020251.99-080136.1 5.98 ±0.04 1.00 0.17 ±0.005 1.11 ±0.01 0.95 ±0.01 0.82 ±0.01 1.10 SDSS J033021.75+001547.1 4.14 ±0.02 1.00 0.64 ±0.005 1.185 SDSS J072403.09+404833.5 4.39 ±0.06 1.00 0.84 ±0.01 1.221 SDSS J075117.08+324425.1 5.16 ±0.06 1.00 0.38 ±0.005 1.39 ±0.02 0.34 ±0.04 0.76 ±0.02 1.12 SDSS J080217.94+112535.0 2.00 ±0.01 1.00 0.93 ±0.01 1.085 SDSS J080441.34+454715.6 3.66 ±0.02 1.00 0.73 ±0.005 1.089 SDSS J082919.82+061744.8 2.97 ±0.01 1.00 0.54 ±0.005 1.247 SDSS J083639.67+471515.3 3.27 ±0.03 1.00 0.46 ±0.005 1.024 SDSS J084251.31+525530.0 6.22 ±0.12 1.00 0.45 ±0.005 1.70 ±0.03 0.62 ±0.04 0.70 ±0.01 0.90 SDSS J084434.40+465214.0 7.68 ±0.30 1.00 0.24 ±0.01 2.13 ±0.04 0.74 ±0.02 0.61 ±0.01 1.08 SDSS J085640.72+055235.4 3.93 ±0.05 1.00 0.53 ±0.01 1.033 SDSS J090222.84+143130.8 14.23 ±0.15 1.00 0.28 ±0.005 2.66 ±0.01 0.82 ±0.005 0.53 ±0.005 1.42 SDSS J093159.95+512254.0 3.91 ±0.07 1.00 0.65 ±0.01 1.28 ±0.03 1.09 ±0.02 0.85 ±0.01 1.01 SDSS J094058.94+400211.3 5.40 ±0.06 1.00 0.19 ±0.005 0.81 ±0.01 0.33 ±0.02 0.85 ±0.01 1.05 CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 39 Table 3.2 (continued) Name red[kpc] ndb/adreb[kpc] nbb/abχ2 ν SDSS J094208.40+094355.5 3.26 ±0.02 1.00 0.45 ±0.005 1.096 SDSS J095146.53+273245.8 2.24 ±0.01 1.00 0.83 ±0.005 1.569 SDSS J095517.41+174114.7 2.47 ±0.01 1.00 0.51 ±0.005 1.171 SDSS J100441.71+282124.0 7.58 ±0.06 1.00 0.21 ±0.005 0.88 ±0.01 0.54 ±0.03 0.90 ±0.01 1.05 SDSS J101422.68+182650.6 2.17 ±0.01 1.00 0.61 ±0.005 1.070 SDSS J102034.04+075106.5 5.47 ±0.08 1.00 0.19 ±0.005 1.04 ±0.01 0.81 ±0.02 0.90 ±0.01 1.02 SDSS J103422.30+442349.1 2.09 ±0.01 1.00 0.58 ±0.005 1.098 SDSS J103543.35+121518.1 8.84 ±0.15 1.00 0.30 ±0.005 1.96 ±0.02 0.73 ±0.02 0.58 ±0.005 1.03 SDSS J103856.94+254521.9 4.81 ±0.05 1.00 0.80 ±0.01 0.891 SDSS J103957.42+174019.5 4.05 ±0.03 1.00 0.60 ±0.005 1.266 SDSS J105153.17+085147.5 7.20 ±0.12 1.00 0.39 ±0.005 1.86 ±0.03 0.80 ±0.03 0.66 ±0.01 1.37 SDSS J110313.24+074253.7 3.35 ±0.02 1.00 0.77 ±0.005 1.211 SDSS J110635.18+440248.7 2.25 ±0.01 1.00 0.46 ±0.005 2.066 SDSS J111554.17+204438.4 8.71 ±0.10 0.12 ±0.01 0.35 ±0.005 1.97 ±0.01 1.00 0.70 ±0.005 1.21 SDSS J112723.89+193849.3 2.22 ±0.01 1.00 0.42 ±0.005 1.084 SDSS J113751.63+215827.1 12.86 ±0.13 1.00 0.61 ±0.005 1.71 ±0.005 1.31 ±0.005 0.93 ±0.005 0.41 SDSS J115759.73+250931.4 2.00 ±0.01 1.00 0.48 ±0.005 1.021 SDSS J120547.65+335021.9 3.74 ±0.05 1.00 0.46 ±0.01 1.296 SDSS J123524.35+474120.6 4.58 ±0.07 1.00 0.24 ±0.005 0.92 ±0.01 0.86 ±0.02 0.92 ±0.01 1.09 SDSS J125558.86+302149.2 4.28 ±0.02 1.00 0.58 ±0.005 1.333 SDSS J125830.33+634234.2 5.28 ±0.08 1.00 0.17 ±0.005 0.97 ±0.01 0.62 ±0.04 0.89 ±0.01 1.45 SDSS J130643.54+093911.4 2.29 ±0.01 1.00 0.92 ±0.01 1.068 CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 40 Table 3.2 (continued) Name red[kpc] ndb/adreb[kpc] nbb/abχ2 ν SDSS J130830.79+503832.0 2.13 ±0.005 1.00 0.56 ±0.005 1.436 SDSS J131138.97+343811.2 5.35 ±0.04 1.00 0.27 ±0.01 5.49 ±0.06 1.21 ±0.01 0.92 ±0.01 0.99 SDSS J131659.28+074326.4 4.07 ±0.03 1.00 0.75 ±0.005 1.222 SDSS J133600.37+063133.9 2.80 ±0.01 1.00 0.63 ±0.005 1.308 SDSS J133700.57+432532.1 7.24 ±0.05 1.00 0.58 ±0.005 1.058 SDSS J135857.84+581406.9 2.33 ±0.01 1.00 0.91 ±0.005 1.062 SDSS J140547.28+151138.3 3.17 ±0.03 1.00 0.56 ±0.005 1.085 SDSS J140929.47+000837.2 3.59 ±0.02 1.00 0.57 ±0.005 1.023 SDSS J141145.48-005415.4 9.93 ±0.20 1.00 0.17 ±0.005 1.79 ±0.02 0.68 ±0.02 0.77 ±0.01 0.99 SDSS J144322.25+010553.2 4.53 ±0.03 1.00 0.60 ±0.005 1.54 ±0.02 0.58 ±0.02 0.59 ±0.01 1.09 SDSS J144718.19+581333.3 3.64 ±0.02 1.00 0.56 ±0.005 1.70 ±0.01 0.42 ±0.02 0.22 ±0.005 1.12 SDSS J145403.72+182401.5 5.12 ±0.06 1.00 0.40 ±0.005 1.65 ±0.02 0.73 ±0.03 0.55 ±0.01 0.95 SDSS J152557.84+481744.8 4.42 ±0.03 1.00 0.92 ±0.01 2.22 ±0.03 1.46 ±0.03 0.20 ±0.005 1.19 SDSS J153235.75+492302.8 2.89 ±0.02 1.00 0.45 ±0.005 1.217 SDSS J155153.04+271433.6 6.11 ±0.09 1.00 0.35 ±0.005 1.43 ±0.01 0.42 ±0.02 0.58 ±0.01 1.49 SDSS J160813.06+440910.2 4.79 ±0.04 1.00 0.56 ±0.005 1.017 SDSS J161159.99+300251.8 5.30 ±0.04 1.00 0.76 ±0.005 0.766 SDSS J161705.55+112506.4 3.46 ±0.01 1.00 0.55 ±0.005 2.376 SDSS J163855.82+132327.0 3.83 ±0.03 1.00 0.54 ±0.005 1.311 SDSS J165529.67+232307.5 10.46 ±0.07 1.00 0.33 ±0.005 1.12 ±0.02 0.70 ±0.06 0.59 ±0.01 0.99 SDSS J165741.87+335509.7 9.36 ±0.18 1.00 0.14 ±0.005 1.57 ±0.02 1.01 ±0.03 0.77 ±0.01 1.95 SDSS J170630.27+220003.9 4.19 ±0.08 0.37 ±0.03 0.63 ±0.01 1.75 ±0.25 1.97 ±0.18 0.47 ±0.01 1.08 CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 41 Table 3.2 (continued) Name red[kpc] ndb/adreb[kpc] nbb/abχ2 ν SDSS J170714.43+652200.2 3.20 ±0.02 1.00 0.50 ±0.005 1.076 SDSS J170842.02+283118.1 4.61 ±0.01 1.00 0.47 ±0.005 1.210 SDSS J213058.11-070507.8 6.23 ±0.04 1.00 0.31 ±0.005 1.16 ±0.01 0.74 ±0.01 0.60 ±0.005 1.78 SDSS J221917.33-011113.7 2.58 ±0.03 1.00 0.53 ±0.005 1.767 CHAPTER 3. STRUCTURE ANALYSIS OF THE SAMPLE GALAXIES 42 Table 3.3: Results of the structural analysis done with GALFIT for the bulgy control sample. Name re [kpc] n b/a χ2 ν SDSS J034357.48+002616 4.16 ±0.06 5.22 ±0.06 0.55 ±0.01 1.002 SDSS J035041.93+010226.7 3.62 ±0.06 5.56 ±0.06 0.80 ±0.01 1.080 SDSS J101345.16-005238 4.67 ±0.22 5.36 ±0.14 0.93 ±0.01 1.071 SDSS J112805.92+000755.9 1.75 ±0.02 3.34 ±0.07 0.72 ±0.01 1.027 SDSS J113029.32+002939.4 6.26 ±0.15 5.62 ±0.07 0.82 ±0.01 1.043 SDSS J121104.06+005820.2 22.11 ±0.44 6.10 ±0.04 0.92 ±0.01 1.031 SDSS J124601.38-010423.5 3.64 ±0.11 5.56 ±0.11 0.88 ±0.01 1.084 SDSS J133708+002707.6 3.83 ±0.07 5.22 ±0.06 0.96 ±0.01 1.120 SDSS J140553.92-004443.5 4.44 ±0.08 5.07 ±0.07 0.69 ±0.01 0.920 SDSS J142959.62+001201.2 2.20 ±0.05 4.81 ±0.10 0.89 ±0.01 0.933 SDSS J145002.16+003443.5 4.99 ±0.13 7.84 ±0.10 0.86 ±0.01 1.034 SDSS J145349.06+000522.7 3.60 ±0.08 6.52 ±0.09 0.76 ±0.01 1.146 SDSS J150444.18-002107.1 7.30 ±0.13 4.83 ±0.05 0.70 ±0.01 1.236 SDSS J155627.54+000333.3 4.58 ±0.17 5.92 ±0.12 0.64 ±0.01 1.015 SDSS J155933.09-010556.1 2.72 ±0.05 3.68 ±0.06 0.71 ±0.01 1.030 SDSS J160453.4-000250.8 2.69 ±0.06 4.80 ±0.09 0.69 ±0.01 0.926 SDSS J160553.82-003301.6 3.35 ±0.07 5.05 ±0.08 0.84 ±0.01 1.100 SDSS J161220.57+004817.5 6.99 ±0.33 6.16 ±0.15 0.81 ±0.01 1.108 SDSS J161757.04-002253.2 4.05 ±0.15 6.41 ±0.14 0.84 ±0.01 0.820 SDSS J163019.13+0011060 3.29 ±0.10 5.11 ±0.13 0.76 ±0.01 1.076 Chapter 4 Stellar content of central regions The spectrum of a galaxy in the optical is composed of photons from every galactic component, mainly star light, which convey information about its general properties. The fraction of blue and red stars, the mass of stars needed to produce such output, the fraction of metals in those stars and many other properties can be deduced by a thorough study. And from that information we can gain insight about its past history such as star-formation or chemical evolution. Turning an observed spectrum into a set of physical properties is not a trivial process. There are two different approaches to solve this problem. One relies on models that reproduce the evolution in time of a composite stellar system with combinations of a stellar evolution prescription with a stellar spectra library [9], the other uses empirical information of individual stars or chemically homogeneous groups of stars of different ages and tries to mimic the observed spectrum using a linear combination of those simpler systems [13]. During this thesis a collaboration was initiated with J. M. Gomes who applied the code STARLIGHT 1[12] - a model of the second type - to the SDSS spectra of the galaxies in the samples analyzed in the previous chapter: galaxies with only disk, with pseudobulges and bulgy ones. This chapter summarizes the results from STARLIGT and the analysis that I performed on those results that aim at giving hints on the different properties and evolutionary path of the central regions of bulgeless and pseudo-bulge galaxies and to compare to those of bulgy galaxies. Note that all quantities derived from STARLIGHT refer only to the central 3” of the 1http://www.starlight.ufsc.br/ 43 CHAPTER 4. STELLAR CONTENT OF CENTRAL REGIONS 44 galaxy (ranging from 1.20 kpc for the closer sources to 3.43 kpc for the farther ones). 4.1 STARLIGHT The basic performance of STARLIGHT is to fit an observed spectrum, Oλ, with a linear combination of N?simple stellar populations (SSP) from evolutionary synthesis models. In this work we used the Bruzual & Charlot (2003) [9] models. Extinction is modeled as due to intervening dust contained in the observed galaxy, and is parametrized by the V-band extinction, AV, using the Galactic extinction law proposed by Cardelli, Clayton & Mathis (1989) [10] with RV≡AV/E(B−V) = 3.1 (where E(B-V) is the color excess between the B-band and the V-band). Kinematic motions in the line of sight are modeled by a Gaussian function, G, centered at velocity v?and with dispersion σ?. With these assumptions, the model spectrum is given by: Mλ=Mλ0 N? X i=1 xibλ,irλ!⊗G(v?, σ?) (4.1) where Mλ0is the synthetic flux at the normalization wavelength λ0= 4020 ˚ A(in this work), xiis the fractional contribution of the SSP with age tiand metallicity Zito the model flux at λ0,bλ,i is the spectrum of the ith SSP normalized at λ0, log10 rλ≡ −0.4(Aλ−Aλ0) is the term for extinction and ⊗is the convolution operator. The base component xican also be denoted by its mass fraction µi. The best fit is chosen by minimizing the χ2, defined as: χ2=X λ [(Oλ−Mλ)ωλ]2(4.2) where ωλis the inverse of the error of the observed spectrum at λ. This definition becomes useful to mask out regions around emission lines, bad pixels and sky residuals by simply setting ωλ= 0. For this study the code ran with a library composed of 25 stellar ages ranging from 106to 1.8×1010 years and with six different metallicities - Zi={0.0001, 0.0004, 0.004, 0.008, 0.02, 0.05}with Z= 0.02. This generates a total of N?= 150 different stellar populations from which the synthetic spectra will be constructed. CHAPTER 4. STELLAR CONTENT OF CENTRAL REGIONS 45 4.2 Inferred physical properties From the fitted spectrum (see Figure 4.1 for two examples) one can infer some physical properties of the underlying stellar population of the galaxies. Some of these properties are rather straightforward to obtain since they play a role in the process of finding the best combination to reproduce the observed spectrum. Extinction and velocity dispersion are parameters directly retrieved from the best model. Stellar mass is not explicitly given but is calculated using the mass fraction, µ, of all components combined with the mass-to-light ratio (M?/Lλ0) characteristic of each SSP. To better quantify the stellar content of the observed regions, the SSPs have been divided in three groups according to their age, •Young population stars with ages t?<9×107years; •Intermediate population stars with 9 ×107< t?<109years; •Old population stars with t?>109years; From this, one can construct six quantities (light fraction and mass fraction of the three defined populations) which quantify the contribution of stars with different ages to the observed spectrum. The mean stellar age is computed as a linear combination of the different SSP’s ages causing the final result to be in the range of ages provided by the library. The lightweighted age is given by loght?iL= N? X i=1 xilog ti(4.3) or alternatively, the mass-weighted age, loght?iM= N? X i=1 µilog ti(4.4) The light-weighted stellar age is affected by any recent star-formation history, as young, blue stars emit strongly in the blue optical part and that will bias the mean age towards lower values. On the other and, the mass-weighted stellar age is biased towards higher mean ages as the older stellar population contributes the most for the CHAPTER 4. STELLAR CONTENT OF CENTRAL REGIONS 52 Figure 4.7: Stellar mass and velocity dispersion for the central regions of the galaxies in the sample. Bulgeless - triangles, pseudo-bulge - open circles, bulgy - filled circles. Figure 4.8: Normalized histogram for the V-band extinction of the central regions of bulgeless (solid line), pseudo-bulge (dashed line) and bulgy (dotted line) galaxies. Chapter 5 Summary, Discussion & Conclusions From an original sample of 77 galaxies issued from the SDSS DR7 through a particular selection scheme I performed a case-by-case structural analysis using GALFIT. 67 galaxies were retained based on their structural properties, and divided in two groups: 38 bulgeless galaxies and 29 pseudo-bulge galaxies. An additional set of 20 bulgy galaxies was also selected to be analyzed in the same way and thus serve as a control sample. The results from a fit to the observed SDSS optical spectrum done with the STARLIGHT spectral synthesis code were further used. These consisted in parameters describing some physical properties such as mean stellar age, stellar mass, metallicity, dust extinction and velocity dispersion issued from the central 3 arcseconds of each galaxy. Then I established comparisons between the derived structure of the galaxies and the parameters computed by STARLIGHT to look for hints on the evolution scenarios for the central regions of these galaxies. 5.1 Discussion & Conclusions In the aftermath of this work one may conclude that: - Distinction between bulgeless and pseudo-bulge galaxies is not trivial but twodimensional structural modeling of galaxies images yields better, more robust, results than fitting of their 1-D surface brightness profiles. 53 CHAPTER 5. SUMMARY, DISCUSSION & CONCLUSIONS 54 - One simple S´ersic profile, when applied on the 1-D surface brightness profile, does not allow us to clearly distinguish between bulgeless and pseudo-bulge galaxies. - The inclination parameter, qam, is not a good probe for galaxy inclination and (b/a)1D is slightly overestimated when considering pseudo-bulge galaxies. - The exponential disks of pseudo-bulge galaxies seem to be larger than those of bulgeless galaxies despite having similar absolute magnitudes which indicates a lower surface brightness for the first ones. - Bulgy galaxies have different physical properties in the central regions when compared with bulgeless and pseudo-bulge galaxies, having higher stellar masses and higher velocity dispersions, having higher mean metallicities and older mean stellar ages and finally by showing almost no star formation in the past 5Gyrs and displaying lower values for dust extinction. - Pseudo-bulge galaxies have higher central stellar masses, higher central velocity dispersions, slightly older mean stellar population ages and lower mass fractions of stars formed in the past 5 Gyrs when compared to the bulgeless sample galaxies. As for metallicity and extinction both these two populations show similar results. Regarding the stellar masses of the central regions of these galaxies one could say that the observed trends are a consequence of the distribution of the mass, as galaxies with a bulge component have their mass more concentrated than bulgeless ones and so, when analyzing this result, one only sees the effect of that concentration. The higher central stellar mass of bulgy galaxies is reasonably explained by mass concentration alone. However, pseudo-bulge galaxies have their disks more extended than the disks of bulgeless galaxies, implying a lower surface brightness of the disk. Assuming that pseudo-bulges are the result of the build up of the central mass on a bulgeless galaxy, either by secular evolution or merger processes, we may ask if the lower surface brightness is related to this build up of mass. If so, this scenario may favor the models of pseudo-bulge growth via satellite accretion. Numerical studies by ElicheMoral et al. (2006) [23] and Aguerri et al. (2001) [2] show that these events lead to an increase of the disk scale length that depends on the mass of the satellite. This occurs due to the outward transport of disk material in the outer regions, combined with inward transport to the bulge in the inner regions. The broadening of the observed spectral lines is due to random motions and/or a coherent rotation. Since our sample galaxies are close to face-on, even if rotation dominates in the observed regions, the contribution to this broadening should not CHAPTER 5. SUMMARY, DISCUSSION & CONCLUSIONS 55 be significant, so we are mainly assessing the magnitude of stellar random motions. Additionally galaxies supported by rotation (disks) tend to have smaller values of thus velocity dispersion than galaxies supported by random motion (ellipticals) [7]. And bulges were found to correlate well with elliptical galaxies as the motion of their stars are mainly random. Thus, as bulges components grow so does the central stellar mass and the corresponding velocity dispersion. The difference in metallicity may be explained also by the observed mass difference as a possible explanation is that since bulgy galaxies in this sample are typically more massive than the disk-dominated galaxies, their gravitational potential should be more effective in retaining the metals expelled in the explosions of Supernovae. These metals thus remain available in the surrounding inter stellar medium to enrich the next (local) generation of stars. If so, even though bulgy galaxies present older stellar populations in average, their mean metallicity can be larger because a significant fraction of their stars incorporated metals that in shallower potentials would have been lost to the disk, halo or even removed from the galaxy. Another possibility is related to the typical environments in which these galaxies have evolved. Works have shown that the metallicity of the galaxies is higher in denser regions when compared to that of the same morphological type but located in low density regions [24] due to tidally triggered star-formation (that enriches more rapidly the inter stellar medium of the galaxies involved). And bulge-dominated galaxies are more characteristic of dense environments such as cluster whereas disk-dominated galaxies tend to live in the field or in the outer regions of clusters [48]. As for the results concerning the observed mean stellar age, taken together with the values obtained for the mass fraction of stars formed in the last 5 Gyrs, the distributions presented in chapter 4 seem to favor the hypothesis that massive bulges form essentially through major merger processes [63] while minor mergers and accretion of satellites are an alternative preferable explanation to the secular formation of pseudobulges [47, Chapter 13]. As for bulgeless galaxies, my results on this parameters seem to support an evolution that is thought to be rather dominated by secular processes (since mergers, even from small satellites, cause a build up of the central mass [23]). The distributions for bulgy galaxies could be explained if major merger processes (or even minor dry mergers) occurred in the early stages of their evolution, depleting the galaxies of their gas and preventing further star formation (or by adding an old stellar population to the galaxy from the accreted satellite, at any time of the galaxy’s life, in the case of dry mergers) leaving the galaxy with an old stellar population. CHAPTER 5. SUMMARY, DISCUSSION & CONCLUSIONS 56 A slow, secular-like evolution scenario in the case of bulgeless galaxies would correspond to a star formation history more extended in time, maintaining significant levels of star formation recently, and a lower mean stellar age. The slightly different distributions observed for the pseudo-bulge galaxies might be explained by the occurrence of minor, gas-rich mergers at a given epoch that locally, and in a shor period of time, accelerated the star formation rate, contributing to the formation of the pseudo-bulge and leaving less gas for later stage star formation. However, the difference between bulgeless and pseudo-bulge stars is not so significant that allows us to distinguish between the two possible scenarios for the formation of pseudo-bulges. Moreover, we are dealing with small numbers so the observed trends can only give indications. If lower extinction values were observed in pseudo-bulges (see section 4.3), this could further favor the hypothesis of pseudo-bulge formation via minor mergers (in detriment of the hypothesis of formation over secular evolution), as this type of interactions can more likely deplete the galaxies of some of their gas and dust. All this results seem to point to the formation of the pseudo-bulges of our sample via minor mergers though one cannot exclude the secular evolution hypothesis. 5.2 Future work Following the purpose of the original study, [15], in the future, higher resolution spectroscopy would allow estimates of the masses of possible central super massive black holes which allow to test whether bulgeless galaxies have these objects in their centers and also to check if the properties of the pseudo-bulges correlate with the black hole mass as happens with classical bulges, contributing to the work of Kormendy et al. (2011) [40]. Additionally, spectroscopic information of the galactic disks of the pseudo-bulge galaxies would allow us to compare the stellar populations of the two components of the galaxy giving further hints on the possible formation scenario of the pseudo-bulge. Appendix A Excluded objects and bulgy sample 57 APPENDIX A. EXCLUDED OBJECTS AND BULGY SAMPLE 58 Figure A.1: Color (gri bands) images of the 36 galaxies excluded from the original sample. APPENDIX A. EXCLUDED OBJECTS AND BULGY SAMPLE 59 Color (gri bands) images of the 36 galaxies excluded from the original sample. (continued) APPENDIX A. EXCLUDED OBJECTS AND BULGY SAMPLE 60 Figure A.2: Color (gri bands) images of the 20 bulgy galaxies. Appendix B GALFIT input file 61 APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES68 Same as Figure 3.9 but for galaxy SDSS J084251.31+525530.0. Same as Figure 3.9 but for galaxy SDSS J084434.40+465214.0. Same as Figure 3.7 but for galaxy SDSS J085640.72+055235.4. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES69 Same as Figure 3.9 but for galaxy SDSS J090222.84+143130.8. Same as Figure 3.9 but for galaxy SDSS J093159.95+512254.0. Same as Figure 3.9 but for galaxy SDSS J094058.94+400211.3. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES70 Same as Figure 3.7 but for galaxy SDSS J094208.40+094355.5. Same as Figure 3.7 but for galaxy SDSS J095146.53+273245.8. Same as Figure 3.7 but for galaxy SDSS J095517.41+174114.7. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES71 Same as Figure 3.9 but for galaxy SDSS J100441.71+282124.0. Same as Figure 3.7 but for galaxy SDSS J101422.68+182650.6. Same as Figure 3.9 but for galaxy SDSS J102034.04+075106.5. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES72 Same as Figure 3.7 but for galaxy SDSS J103422.30+442349.1. Same as Figure 3.9 but for galaxy SDSS J103543.35+121518.1. Same as Figure 3.7 but for galaxy SDSS J103856.94+254521.9. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES73 Same as Figure 3.7 but for galaxy SDSS J103957.42+174019.5. Same as Figure 3.9 but for galaxy SDSS J105153.17+085147.5. Same as Figure 3.7 but for galaxy SDSS J110313.24+074253.7. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES74 Same as Figure 3.7 but for galaxy SDSS J110635.18+440248.7. Same as Figure 3.9 but for galaxy SDSS J111554.17+204438.4. Same as Figure 3.7 but for galaxy SDSS J112723.89+193849.3. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES75 Same as Figure 3.9 but for galaxy SDSS J113751.63+215827.1. Same as Figure 3.7 but for galaxy SDSS J115759.73+250931.4. Same as Figure 3.7 but for galaxy SDSS J120547.65+335021.9. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES76 Same as Figure 3.9 but for galaxy SDSS J123524.35+474120.6. Same as Figure 3.7 but for galaxy SDSS J125558.86+302149.2. Same as Figure 3.9 but for galaxy SDSS J125830.33+634234.2. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES77 Same as Figure 3.7 but for galaxy SDSS J130643.54+093911.4. Same as Figure 3.7 but for galaxy SDSS J130830.79+503832.0. Same as Figure 3.9 but for galaxy SDSS J131138.97+343811.2. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES84 Same as Figure 3.9 but for galaxy SDSS J165741.87+335509.7. Same as Figure 3.9 but for galaxy SDSS J170630.27+220003.9. Same as Figure 3.7 but for galaxy SDSS J170714.43+652200.2. APPENDIX C. GALFIT RESIDUALS AND SURFACE BRIGHTNESS PROFILES85 Same as Figure 3.7 but for galaxy SDSS J170842.02+283118.1. Same as Figure 3.9 but for galaxy SDSS J213058.11-070507.8. Same as Figure 3.7 but for galaxy SDSS J221917.33-011113.7. References [1] K. N. Abazajian, J. K. Adelman-McCarthy, M. A. Ag¨ueros, S. S. Allam, C. Allende Prieto, D. An, K. S. J. Anderson, S. F. Anderson, J. Annis, N. A. Bahcall, and et al. The Seventh Data Release of the Sloan Digital Sky Survey. ApJS, 182:543–558, June 2009. [2] J. A. L. Aguerri, M. Balcells, and R. F. Peletier. Growth of galactic bulges by mergers. I. Dense satellites. A&A, 367:428–442, Feb. 2001. [3] E. Athanassoula. On the nature of bulges in general and of box/peanut bulges in particular: input from N-body simulations. MNRAS, 358:1477–1488, Apr. 2005. [4] M. Balcells and R. F. Peletier. Colors and color gradients in bulges of galaxies. AJ, 107:135–152, Jan. 1994. [5] J. C. Barentine and J. Kormendy. Two Pseudobulges in the ”Boxy Bulge” Galaxy NGC 5746. ApJ, 754:140, Aug. 2012. [6] E. F. Bell. Galaxy Bulges and their Black Holes: a Requirement for the Quenching of Star Formation. ApJ, 682:355–360, July 2008. [7] J. Binney. Dynamics of elliptical galaxies and other spheroidal components. ARA&A, 20:399–429, 1982. [8] M. R. Blanton, D. J. Schlegel, M. A. Strauss, J. Brinkmann, D. Finkbeiner, M. Fukugita, J. E. Gunn, D. W. Hogg, ˇ Z. Ivezi´c, G. R. Knapp, R. H. Lupton, J. A. Munn, D. P. Schneider, M. Tegmark, and I. Zehavi. New York University Value-Added Galaxy Catalog: A Galaxy Catalog Based on New Public Surveys. AJ, 129:2562–2578, June 2005. [9] G. Bruzual and S. Charlot. Stellar population synthesis at the resolution of 2003. MNRAS, 344:1000–1028, Oct. 2003. 86 REFERENCES 87 [10] J. A. Cardelli, G. C. Clayton, and J. S. Mathis. The relationship between infrared, optical, and ultraviolet extinction. ApJ, 345:245–256, Oct. 1989. [11] B. Carroll and D. Ostlie. An introduction to modern astrophysics, ch. 25. Pearson Addison-Wesley, 2007. [12] R. Cid Fernandes, A. Mateus, L. Sodr´e, G. Stasi´nska, and J. M. Gomes. Semiempirical analysis of Sloan Digital Sky Survey galaxies - I. Spectral synthesis method. MNRAS, 358:363–378, Apr. 2005. [13] R. Cid Fernandes, L. Sodr´e, H. R. Schmitt, and J. R. S. Le˜ao. A probabilistic formulation for empirical population synthesis: sampling methods and tests. MNRAS, 325:60–76, July 2001. [14] L. Ciotti and G. Bertin. Analytical properties of the R1/m law. A&A, 352:447–451, Dec. 1999. [15] B. Coelho. AGN feedback and quenching of star formation: a multiwavelength approach with the EURO-VO. Master’s thesis, Universidade do Porto, Portugal, 2010. [16] C. J. Conselice. The Relationship between Stellar Light Distributions of Galaxies and Their Formation Histories. ApJS, 147:1–28, July 2003. [17] J. J. Dalcanton, P. Yoachim, and R. A. Bernstein. The Formation of Dust Lanes: Implications for Galaxy Evolution. ApJ, 608:189–207, June 2004. [18] W. J. G. de Blok, J. M. van der Hulst, and G. D. Bothun. Surface photometry of low surface brightness galaxies. MNRAS, 274:235–255, May 1995. [19] V. de Lapparent, A. Baillard, and E. Bertin. The EFIGI catalogue of 4458 nearby galaxies with morphology. II. Statistical properties along the Hubble sequence. A&A, 532:A75, Aug. 2011. [20] G. de Vaucouleurs. Recherches sur les Nebuleuses Extragalactiques. Annales d’Astrophysique, 11:247, Jan. 1948. [21] G. de Vaucouleurs. Classification and Morphology of External Galaxies. Handbuch der Physik, 53:275, 1959. [22] S. Djorgovski and M. Davis. Fundamental properties of elliptical galaxies. ApJ, 313:59–68, Feb. 1987. REFERENCES 88 [23] M. C. Eliche-Moral, M. Balcells, J. A. L. Aguerri, and A. C. Gonz´alez-Garc´ıa. Growth of galactic bulges by mergers. II. Low-density satellites. A&A, 457:91– 108, Oct. 2006. [24] S. L. Ellison, L. Simard, N. B. Cowan, I. K. Baldry, D. R. Patton, and A. W. McConnachie. The mass-metallicity relation in galaxy clusters: the relative importance of cluster membership versus local environment. MNRAS, 396:1257– 1272, July 2009. [25] S. M. Faber and R. E. Jackson. Velocity dispersions and mass-to-light ratios for elliptical galaxies. ApJ, 204:668–683, Mar. 1976. [26] D. B. Fisher and N. Drory. The Structure of Classical Bulges and Pseudobulges: the Link Between Pseudobulges and S´ ERSIC Index. AJ, 136:773–839, Aug. 2008. [27] D. B. Fisher and N. Drory. Bulges of Nearby Galaxies with Spitzer: Scaling Relations in Pseudobulges and Classical Bulges. ApJ, 716:942–969, June 2010. [28] K. C. Freeman. On the Disks of Spiral and so Galaxies. ApJ, 160:811, June 1970. [29] M. Fukugita, T. Ichikawa, J. E. Gunn, M. Doi, K. Shimasaku, and D. P. Schneider. The Sloan Digital Sky Survey Photometric System. AJ, 111:1748, Apr. 1996. [30] D. A. Gadotti. Structural properties of pseudo-bulges, classical bulges and elliptical galaxies: a Sloan Digital Sky Survey perspective. MNRAS, 393:1531– 1552, Mar. 2009. [31] Y. Guo, D. H. McIntosh, H. J. Mo, N. Katz, F. C. van den Bosch, M. Weinberg, S. M. Weinmann, A. Pasquali, and X. Yang. Structural properties of central galaxies in groups and clusters. MNRAS, 398:1129–1149, Sept. 2009. [32] E. P. Hubble. Extragalactic nebulae. ApJ, 64:321–369, Dec. 1926. [33] P. Jablonka, P. Martin, and N. Arimoto. The Luminosity-Metallicity Relation for Bulges of Spiral Galaxies. AJ, 112:1415, Oct. 1996. [34] R. I. Jedrzejewski. CCD surface photometry of elliptical galaxies. I - Observations, reduction and results. MNRAS, 226:747–768, June 1987. [35] W. A. Joye and E. Mandel. New Features of SAOImage DS9. In H. E. Payne, R. I. Jedrzejewski, and R. N. Hook, editors, Astronomical Data Analysis Software and Systems XII, volume 295 of Astronomical Society of the Pacific Conference Series, page 489, 2003. REFERENCES 89 [36] S. J. Kautsch. The Edge-On Perspective of Bulgeless, Simple Disk Galaxies. PASP, 121:1297–1306, Dec. 2009. [37] J. A. Keselman and A. Nusser. Pseudo-bulge formation via major mergers. MNRAS, 424:1232–1243, Aug. 2012. [38] S. Khochfar. Merger History of Galaxies and Disk+Bulge Formation. In S. Jogee, I. Marinova, L. Hao, and G. A. Blanc, editors, Galaxy Evolution: Emerging Insights and Future Challenges, volume 419 of Astronomical Society of the Pacific Conference Series, page 197, Dec. 2009. [39] J. Kormendy. Observations of galaxy structure and dynamics. In L. Martinet and M. Mayor, editors, Saas-Fee Advanced Course 12: Morphology and Dynamics of Galaxies, pages 113–288, 1982. [40] J. Kormendy, R. Bender, and M. E. Cornell. Supermassive black holes do not correlate with galaxy disks or pseudobulges. Nature, 469:374–376, Jan. 2011. [41] J. Kormendy, N. Drory, R. Bender, and M. E. Cornell. Bulgeless Giant Galaxies Challenge Our Picture of Galaxy Formation by Hierarchical Clustering. ApJ, 723:54–80, Nov. 2010. [42] J. Kormendy and R. C. Kennicutt, Jr. Secular Evolution and the Formation of Pseudobulges in Disk Galaxies. ARA&A, 42:603–683, Sept. 2004. [43] M. Kregel and P. C. van der Kruit. Radial truncations in stellar discs in galaxies. MNRAS, 355:143–146, Nov. 2004. [44] D. Larson, J. Dunkley, G. Hinshaw, E. Komatsu, M. R. Nolta, C. L. Bennett, B. Gold, M. Halpern, R. S. Hill, N. Jarosik, A. Kogut, M. Limon, S. S. Meyer, N. Odegard, L. Page, K. M. Smith, D. N. Spergel, G. S. Tucker, J. L. Weiland, E. Wollack, and E. L. Wright. Seven-year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Power Spectra and WMAP-derived Parameters. ApJS, 192:16, Feb. 2011. [45] L. D. Matthews and J. S. Gallagher, III. B and V CCD Photometry of Southern, Extreme Late-Type Spiral Galaxies. AJ, 114:1899, Nov. 1997. [46] L. D. Matthews and J. M. Uson. H I Imaging Observations of Superthin Galaxies. II. IC 2233 and the Blue Compact Dwarf NGC 2537. AJ, 135:291–318, Jan. 2008. REFERENCES 90 [47] H. Mo, F. C. van den Bosch, and S. White. Galaxy Formation and Evolution. 2010. [48] A. Oemler, Jr. The Systematic Properties of Clusters of Galaxies. Photometry of 15 Clusters. ApJ, 194:1–20, Nov. 1974. [49] C. Y. Peng, L. C. Ho, C. D. Impey, and H.-W. Rix. Detailed Structural Decomposition of Galaxy Images. AJ, 124:266–293, July 2002. [50] C. Y. Peng, L. C. Ho, C. D. Impey, and H.-W. Rix. Detailed Decomposition of Galaxy Images. II. Beyond Axisymmetric Models. AJ, 139:2097–2129, June 2010. [51] M. Pohlen and I. Trujillo. The structure of galactic disks. Studying late-type spiral galaxies using SDSS. A&A, 454:759–772, Aug. 2006. [52] W. H. Press, S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery. Numerical Recipes 3rd Edition: The Art of Scientific Computing. Cambridge University Press, New York, NY, USA, 3 edition, 2007. [53] C. W. Purcell, S. Kazantzidis, and J. S. Bullock. Galactic Disk Transformation via Massive Satellite Accretion Events. In S. Jogee, I. Marinova, L. Hao, and G. A. Blanc, editors, Galaxy Evolution: Emerging Insights and Future Challenges, volume 419 of Astronomical Society of the Pacific Conference Series, page 248, Dec. 2009. [54] R. Roˇskar, V. P. Debattista, G. S. Stinson, T. R. Quinn, T. Kaufmann, and J. Wadsley. Beyond Inside-Out Growth: Formation and Evolution of Disk Outskirts. ApJL, 675:L65–L68, Mar. 2008. [55] J. L. Sersic. Atlas de galaxias australes. 1968. [56] L. Sparke and I. John S. Gallagher. Galaxies in the Universe: An Introduction. Cambridge University Press, 2007. [57] S. van den Bergh. A Preliminary Luminosity Clssification of Late-Type Galaxies. ApJ, 131:215, Jan. 1960. [58] F. C. van den Bosch. The Formation of Disk-Bulge-Halo Systems and the Origin of the Hubble Sequence. ApJ, 507:601–614, Nov. 1998. [59] P. C. van der Kruit. Optical surface photometry of eight spiral galaxies studied in Westerbork. A&AS, 38:15–38, Oct. 1979. REFERENCES 91 [60] R. A. Vincent and B. S. Ryden. The Dependence of Galaxy Shape on Luminosity and Surface Brightness Profile. ApJ, 623:137–147, Apr. 2005. [61] Y. Wadadekar, B. Robbason, and A. Kembhavi. Two-dimensional Galaxy Image Decomposition. AJj, 117:1219–1228, Mar. 1999. [62] D. G. York, J. Adelman, J. E. Anderson, Jr., S. F. Anderson, J. Annis, N. A. Bahcall, J. A. Bakken, R. Barkhouser, S. Bastian, E. Berman, W. N. Boroski, S. Bracker, C. Briegel, J. W. Briggs, J. Brinkmann, R. Brunner, S. Burles, L. Carey, M. A. Carr, F. J. Castander, B. Chen, P. L. Colestock, A. J. Connolly, J. H. Crocker, I. Csabai, P. C. Czarapata, J. E. Davis, M. Doi, T. Dombeck, D. Eisenstein, N. Ellman, B. R. Elms, M. L. Evans, X. Fan, G. R. Federwitz, L. Fiscelli, S. Friedman, J. A. Frieman, M. Fukugita, B. Gillespie, J. E. Gunn, V. K. Gurbani, E. de Haas, M. Haldeman, F. H. Harris, J. Hayes, T. M. Heckman, G. S. Hennessy, R. B. Hindsley, S. Holm, D. J. Holmgren, C.-h. Huang, C. Hull, D. Husby, S.-I. Ichikawa, T. Ichikawa, ˇ Z. Ivezi´c, S. Kent, R. S. J. Kim, E. Kinney, M. Klaene, A. N. Kleinman, S. Kleinman, G. R. Knapp, J. Korienek, R. G. Kron, P. Z. Kunszt, D. Q. Lamb, B. Lee, R. F. Leger, S. Limmongkol, C. Lindenmeyer, D. C. Long, C. Loomis, J. Loveday, R. Lucinio, R. H. Lupton, B. MacKinnon, E. J. Mannery, P. M. Mantsch, B. Margon, P. McGehee, T. A. McKay, A. Meiksin, A. Merelli, D. G. Monet, J. A. Munn, V. K. Narayanan, T. Nash, E. Neilsen, R. Neswold, H. J. Newberg, R. C. Nichol, T. Nicinski, M. Nonino, N. Okada, S. Okamura, J. P. Ostriker, R. Owen, A. G. Pauls, J. Peoples, R. L. Peterson, D. Petravick, J. R. Pier, A. Pope, R. Pordes, A. Prosapio, R. Rechenmacher, T. R. Quinn, G. T. Richards, M. W. Richmond, C. H. Rivetta, C. M. Rockosi, K. Ruthmansdorfer, D. Sandford, D. J. Schlegel, D. P. Schneider, M. Sekiguchi, G. Sergey, K. Shimasaku, W. A. Siegmund, S. Smee, J. A. Smith, S. Snedden, R. Stone, C. Stoughton, M. A. Strauss, C. Stubbs, M. SubbaRao, A. S. Szalay, I. Szapudi, G. P. Szokoly, A. R. Thakar, C. Tremonti, D. L. Tucker, A. Uomoto, D. Vanden Berk, M. S. Vogeley, P. Waddell, S.-i. Wang, M. Watanabe, D. H. Weinberg, B. Yanny, N. Yasuda, and SDSS Collaboration. The Sloan Digital Sky Survey: Technical Summary. AJ, 120:1579–1587, Sept. 2000. [63] J. Zavala, V. Avila-Reese, C. Firmani, and M. Boylan-Kolchin. The growth of galactic bulges through mergers in LCDM haloes revisited. I. Present-day properties. ArXiv e-prints, Apr. 2012.