There is now considerable interest in how stellar streams in the Milky Way can be used to probe how the earlier merger history of our galaxy, which in turn can be related to hierarchical models of galaxy evolution.

We have used the Hubble Space Telescope's Advanced Camera for Surveys to measure the mass density function of morphologically-selected early-type galaxies in the Gemini Deep Deep Survey fields, over the redshift range 0.9 < z < 1.6. Our imaging data set covers four well-separated sight-lines, and is roughly intermediate (in terms of both depth and area) between the GOODS/GEMS imaging data, and the images obtained in the Hubble Deep Field campaigns. Our images contain 144 galaxies with ultra-deep spectroscopy, and they have been analyzed using a new purpose-written morphological analysis code which improves the reliability of morphological classifications by adopting a ‘quasi-petrosian’ image thresholding technique. We find that at z = 1 approximately 70% of the stars in massive galaxies reside in early-type systems. This fraction is remarkably similar to that seen in the local Universe. However, we detect very rapid evolution in this fraction over the range 1.0 < z < 1.6, suggesting that in this epoch the strong color-morphology relationship seen in the nearby Universe is beginning to fall into place.

The latest observational results on the formation and evolution of early-type galaxies are reviewed by focusing on the issues of mass assembly and downsizing scenario.

The study of the spatial distribution of interstellar extinction, AV, is important for many investigations of galactic and extragalactic objects. Three-dimensional (3D) extinction models have been produced using spectral and photometric stellar data, open cluster data, star counts, the Galactic dust distribution model.

In this paper we describe three lines of work that are being developed in the framework of the Spanish Virtual Observatory.

The NSF Science and Technology Center for Adaptive Optics (CfAO) is supporting a major scientific legacy project called the CfAO Treasury Survey (CATS). CATS is obtaining near-infrared AO data in deep HST survey fields, such as GEMS, GOODS-N, & EGS. Besides summarizing the main objectives of CATS, we highlight some recent imaging work on the study of distant field galaxies, AGNs, and a redshift z = 1.32 supernova. CATS plans the first data release to the community in early 2007 (check http://www.astro.ucla.edu/~irlab/cats/index.shtml for more details on CATS and latest updates).

We have compiled a large number of optical spectra of cool stars taken with different high-resolution echelle spectrographs (R ≃ 40 000). Many of those are available as spectral libraries (Montes et al. 1997, 1998, 1999, <http://www.ucm.es/info/Astrof/invest/actividad/spectra.html)>.

I will discuss the search for galaxies and QSOs at the highest redshifts, out to the Epoch of Reionization. Observations in the submillimeter play a fundamental role in such searches, in particular once ALMA will become operational. I will focus on the properties of the molecular gas (the phase of the ISM out of which stars form) in the earliest systems as derived through observations in the millimetre and radio regime. These observations allow us to measure the reservoir of molecular gas and to constrain the physical properties and excitation of the gas. Resolved imaging of the host galaxies also allows us to derive first dynamical masses for these objects. Such mass determinations are critical to determine whether there is a possible evolution of the famous MBH–σbulge relation with redshift or not. Recent measurements in the QSO redshift record holder J1148+5251 (z= 6.42) provide evidence that the black hole in this system assembled before the stellar bulge.

Current surveys from modern observatories contain a huge amount of information; in particular, the Sloan Digital Sky Survey (SDSS) has reached the order of terabytes of data in images and spectra. Such amount of information needs to be exploited by sophisticated algorithms that automatically analyze the data in order to extract useful knowledge from the mega databases.

We investigate the properties of Lyman Break Galaxies at z ~ 3 in the Spitzer First Look Survey field. The effective area coverage of the study is ~2.63 deg2, the largest among similar studies of R ~ 24 mag u-dropouts. Photometric redshifts, ages, dust extinctions and stellar masses of the LBGs have been derived through SED fitting. More than 70% of IRAC-detected LBGs have stellar masses of order of ~1011 M⊙, comparable with L* galaxies. The MIR luminosity, which represents the rest-frame NIR, is found to be a good tracer of the stellar mass. As the red (optical−MIR) colors of the massive LBGs suggest, IRAC LBGs are older (〈t〉 ~ 0.9 Gyr) and dustier (〈E(B − V)〉 ~ 0.33) than typical LBGs. The age and the amount of dust indicate that the IRAC LBGs are intermediate population between blue LBG population and red DRG population. The number density of LBGs more massive than 1011 M⊙ is Φ = (1.05 ± 0.15) × 10−5 Mpc−3. This value exceeds the predictions of hierarchicl galaxy formation models.

The Virtual Observatory (VO) is often presented as an all-purpose tool for the astronomer to do all his current work in a single web browser, from which he can control all data searches, reductions and mainly the full data analysis. This idea seems to be already fulfilling in certain fields of astronomy like is the search of unknown correlations between various parameters in large scale surveys, galaxy statistics or cross-identification of objects. The current VO tools are, however, lacking support for important capabilities needed in stellar spectroscopy. Especially the high-resolution spectroscopy of stars with rapid line profile variations or of those with complicated emission profiles (like Be or symbiotic stars) benefits from a number of specific methods implemented today only in stand-alone legacy packages working with local data.

The development of large ground-based telescopes and sensitive large format detectors, as well as the develepment of various techniques for the selection of high-z galaxies enabled us to construct large samples of galaxies in the early universe, as reported in the many contributions in this proceedings. The next major step for the comprehensive understanding of the galaxy evolution would be to explore the relationship of galaxies selected with different criteria at different epochs and find links between them. In this contribution we present the properties of Lyman break galaxies (LBGs) at z ~ 5 obtained by deep and wide blank field surveys, and through the comparison with samples at lower redshift ranges we discuss the evolution of star-forming galaxies in the early universe.

We review the current status of cosmological simulations of the formation and evolution of disc galaxies, discussing in particular the Tully-Fisher relation and its redshift evolution.

The term ‘grid’, in the Virtual Observatory (VO) context, has mainly been used to indicate a set of interoperable services, allowing transparent access to a set of geographically distributed and heterogeneous archives and catalogues, data exchange and analysis, etc. The design of the VO has been however mainly geared at allowing users to access registered services.

We search for massive and evolved galaxies at z ≥ 5 in the Great Observatories Origins Deep Survey (GOODS) southern field. Combining HST ACS, VLT ISAAC and Spitzer IRAC broad–band photometric data, we develop a color selection technique to identify candidates for being evolved galaxies at high redshifts. The color selection is primarily based on locating the Balmer–break using the K- and 3.6 μm bands. Stellar population synthesis models are fitted to the SEDs of these galaxies to identify the final sample. We find 11 candidates with photometric redshifts in the range 4.9 < z < 5.6, dominated by an old stellar population, with ages 0.2-1.0 Gyr, and stellar masses in the range (0.7 − 5) × 1011 M⊙. Most of the candidates have modest amounts of internal dust extinction. The majority of the stars in these galaxies were formed at z>9 and the current star formation activity is a few percent of the inferred initial star formation rate.

There exist two principally different approaches to the organization of problem solving in VO: (i) information resources driven approach (choice and integrated definition of resources are made independently of the problem specification); and (ii) scientific problem driven approach (a specification of a problem domain is created, the relevant to the problem resources are identified and semantically mapped into the domain). Intrinsic difficulties of the first approach: semantic gap between resources and the problem, instability of global schema w.r.t. a set of resources, inability of automatic identification of resources for the problem. To implement the second approach a mediation technology is required. On the consolidation phase of the mediator the efforts of the scientific community are focused on the problem definition by specifying the mediator. During the operational phase relevant information resources are identified and expressed in terms of the mediator. Advantages of the mediator approach include truly semantic integration of heterogeneous resources due to their semantic mapping into the mediator; multiple subjects can be semantically integrated applying recursive structure of the mediators.

We have assembled large samples of galaxies at redshift z ~ 4, 5 and 6 (totalling >4300 objects, >1000 objects, >500 objects, respectively) from all the deep HST ACS and NICMOS data taken to date (over 2000 orbits of data). From these we have derived rest-frame UV luminosity functions, luminosity densities, and star formation rates in a very robust and consistent way to very faint luminosities (0.01L* to 0.04L*). The faint-end slopes α of these luminosity functions are remarkably uniform and steep (α ~ −1.7), indicating very little evolution from z ~ 6 to z ~ 4. The characteristic luminosity L* brightens considerably (by ~1 mag) over this period, but the overall change in the luminosity function is such as to lead to little change in the luminosity density and star formation rate over this time. We also have detected galaxies at z ~ 7 − 8 and set strong limits at z ~ 10 directly from deep HST NICMOS observations. Spitzer observations of these z ~ 7 galaxies have been used to estimate masses and ages, suggesting substantial formation at z ~ 10 or earlier. These results show that this hierachical build-up continues into the reionization epoch.

The Semantic Web and ontologies are emerging technologies that enable advanced knowledge management and sharing. Their application to Astronomy can offer new ways of sharing information between astronomers, but also between machines or software components and allow inference engines to perform reasoning on an astronomical knowledge base. The first examples of astronomy-related ontologies are being developed in the european VOTech project.

A Virtual Observatory (VO) seeks to provide access to very large volumes of astronomical data over the internet to every interested user, along with software for the data anlysis, and tools for data visualisation, statistics and any other applications which are necessary for mining science from the data. Handling large volumes of data requires extensive computer resources, which may not be available to the end user; transferring data from the data centre to the user needs high bandwidth, which also may not be available. A VO therefore seeks to provide computing resources as well, which could be spread over a grid. The user can locate data of interest through registries, then access the data and analyse it using the computing resources, all through simple user interfaces.

Clumpy galaxies are prominent in the early Universe. We present morphological and photometric properties of a wide range of galaxy types and their star-forming clumps in the Hubble Ultra Deep Field. Sizes, scale lengths, and scale heights suggest that galaxies grow by a factor of 2 from z = 4 to the present, and that thick disks are present in the early Universe. The largest clumps of star formation are 107–109M⊙ in different galaxies, much more massive than large star-forming complexes in local galaxies. Dissolved clumps may account for both the exponential disks and the early thick disks of spirals and proto-spiral galaxies.