Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/~dejong/thes/chap05.pdf Дата изменения: Tue Dec 18 22:07:25 2007 Дата индексирования: Tue Oct 2 05:35:42 2012 Кодировка: Поисковые слова: http astrokuban.info astrokuban

Chapter 5 Conclusions & prospects
Abstract. In this thesis optical and near-infrared surface photometry has been used to perform a statistical study of light and color distributions in and among spiral galaxies. In this chapter I summarize the main results using the six passband surface photometry of 86 face-on spiral galaxies and give some suggestions for future work.

1 The distribution of light I analyzed the problem of separatin galaxies into different components of decomposition methods were test revealed the two main sources of err g the light distribution of in Chapter 2. A number ed and compared and this o r:

2 The distribution of colors In Chapter 4 I investigated the dust and stellar content of the sample of galaxies using their color gradients. The sample was used as a whole, no individual galaxies were examined. This has some clear advantages. While it is conceivable that an individual galaxy has a stellar population with a metallicity much higher than the metallicity of the surrounding gas due to a recent merging with a metal poor dwarf, this is very unlikely to be the case for most of the galaxies in a sample. The same holds true for extreme dust distributions. While a number of galaxies could have strange dust properties and distributions, it is unlikely that all galaxies have properties and distribution different from what is observed in edge-on galaxies and especially in our own Galaxy. The main conclusions of Chapter 4 are: ­ Almost all spiral galaxies become bluer with increasing radius. ­ The colors of galaxies correlate strongly with surface brightness, with lower surface brightness regions being bluer, both within and among galaxies. The morphological type is an additional parameter in this relationship, because at the same surface brightness late-type galaxies are bluer than early-type galaxies. ­ The color gradients in the galaxies are best explained by differences in SFH as function of radius, with the outer parts of galaxies being much younger on average than the central regions. This implies that the stellar scalelength of galaxies is still growing. The central stellar populations must have a range in metallicities to explain the red central colors of the galaxies. ­ Realistic 3D radiative transfer modeling indicates that reddening due to dust extinction cannot be the major cause of the color gradients in face-on galaxies. The model color vectors in color­color space are not compatible with the data. ­ A consequence of the population changes implied by the color differences in and among galaxies is that there are large differences in M =L values in and among galaxies. These changes in M =L make the missing mass problem in spiral galaxies as derived from rotation curve fitting even more severe. The key figures of Chapter 4 are Figs 2, 6 and 7.

­ The uncertainty in the shape of the bulge profile to be used. ­ The uncertainty in the sky surface brightness. An analysis of the fit residuals shows that the residual variations among the different fit models are small compared to the residuals themselfs. The accuracy of the obtained fundamental parameters can only be improved by increasing the description of the different model components. The most reproducible and representative bulge and disk parameters are obtained by modeling the two-dimensional image of the galaxy with an exponential light profile for both bulge and disk and, when necessary, with a Freeman bar (Freeman 1966) The fundamental bulge and disk parameters were statistically analyzed in Chapter 3. The observed sample is statistically complete and correctable for selection effects. The K passband observations gave a nearly unobscured view on the galaxies revealing the "old" stellar distribution of the galaxies. The main results from this chapter are: ­ Freeman's law (Freeman 1970) of constant disk central surface brightness among galaxies has to be redefined. There is an upper limit to the central surface brightnesses as indicated by Freeman's law, while for lower central surface brightnesses the number density of galaxies decreases only slowly as function of central surface brightness. ­ The bivariate distributions in the ( 0 ,h)-plane (Fig. 7) indicate that there might be large numbers of small scalelength, low surface brightness galaxies, but this area in the diagram is poorly sampled. ­ The Hubble sequence type index correlates strongly with the effective surface brightness of the bulge, much better than with the bulge-to-disk ratio. ­ The disk and bulge scalelengths are correlated. ­ These scalelengths are not correlated with Hubble type. Hubble type is a lengthscale-free parameter and each type therefore comes in a range of magnitudes (and presumably a range of total masses). The key figures of Chapter 3 are Figs 5, 7, 8 and 10.


134

j

CH

APTER

5

j

CO

NCLUSIONS

&

PROSPECTS

3 Future work Every study raises more questions than it solves. Some suggestions for future work and observational programs are presented in this final section. Two problems regarding Freeman's law were not dealt with adequately in the present study. 1) The number density of low surface brightness and small scalelength galaxies is still unknown, as figure 5 of Chapter 3 shows that the present sample is heavily biased against such galaxies. 2) The galaxies of the present sample are probably not randomly distributed in space. The sample may be slightly biased due to the density enhancement of the local supercluster. Both problems can be circumvented by observing a new sample using careful selection criteria. The new diameter limited sample should be selected from an all sky (or at least half sky) survey from deep photographic plates enabling fainter isophotal diameters to be used (preferably measured and not eye estimated) than the diameters used here. This will shift the selection lines in Fig. 5, Chapter 3 downwards. Using a smaller diameter criterion ( 1­1.5 arcmin) will shift the selection lines to the left. When the sample is limited to only face-on galaxies, it will still contain several thousands of galaxies. By selecting at random one in twenty galaxies one ensures random distribution over the whole sky and combined with the faint limits this ensures that the sample is hardly biased by the local density enhancement. At least several hundreds of galaxies should be observed (surface photometry in the K passband of course, but also their redshifts should be measured) to fill the bivariate distributions, such as Figs 7 and 23, adequately. The distribution of dust and the amount of extinction it is causing can soon be measured much more accurately at the far infrared bands by the use of ISO and at the submillimeter wavelengths by the use of SCUBA at the JCMT. The determination of the scattering properties of dust needs most urgently improvements. The dust scattering properties used in this thesis stem from observations made in the early seventies or from model calculations. Scattering properties of dust in extra-galactic systems have never been measured. The nature of the color gradients can be better determined when spectroscopy is used to determine the stellar population properties as function of radius. A first indication of the metal-

licity gradient can be obtained by measuring a number of H II regions, but this gives only information about the metallicity gradient in the current gas fraction. Ideally one would like to obtain the long-slit spectra of the stellar population of the galaxies, concentrating on the absorption lines indicated by Worthey (1994) to be the most affected by age or metallicity effects. Such observations would require long observation times, especially for the faint outer parts of galaxies and for low surface brightness galaxies. Maybe the new 8-10 m class telescopes which will become available during the next few years will decrease the demand on the 4 m class telescopes, which could be used to obtain long-slit spectra in reasonable observing times of a statistically significant number of galaxies. There is a clear need for a systematic survey of broadband optical and in particular near-infrared surface photometry of the large, nearby galaxies. These galaxies have been studied at all wavelengths in great detail and the local physical processes in these galaxies are known in some detail. It is important to know the relation between these local processes and the global light distribution of these galaxies. All galaxies that can in principle be resolved in stars from the ground (like the Magellanic clouds, M 31, M 33 etc.) should be subject of such a survey and could be observed with a small, wide field telescope (like the parking lot camera of Bothun & Thompson 1988). Once it has been established what fraction of the color gradients is caused by dust, age and metallicity effects, it becomes interesting to examine the physical processes determining the steepness of the color gradient. For this sample no correlation between the steepness of the gradient and any other known galaxy parameter has been found so far. Peletier et al. (1994) showed a very weak correlation with inclination, but this can not be confirmed with this face-on sample. References
Bothun G.D., Thompson I.B. 1988, AJ 96, 877 Freeman K.C. 1966, MNRAS 133, 47 Freeman K.C. 1970, ApJ 160, 811 Peletier R.F., Valentijn E.A., Moorwood A.F.M., Freudling W. 1994, A&AS 108, 621 Worthey G. 1994, ApJS 95, 107