Context. The study of high redshift Tully-Fisher relations (TFRs) is limited by the use of long slit spectrographs, rest frame B band and star formation selected galaxies. Aims. We try to circumvent these issues by using integral field spectroscopy (SINFONI), by studying the rest frame K band and stellar mass TFR, and by selecting targets without a bias to strongly star forming galaxies. In this paper, we demonstrate our methods on our best case. This galaxy, F257, at z=2.03, was selecte from a sample of candidate high redshift large disk galaxies in the Hubble Deep Field South that were selected with photometric and morphological criteria. Methods. We used SINFONI at the VLT to obtain an integral field spectrum of the Halpha line and hence a velocity field and rotation curve. We also use UBVIJHK+IRAC band photometry to determine a stellar photometric mass. Results. We find that F257 is indistinguishable from local late type galaxies in many respects: it has a regular velocity field, increasing velocity disperion towards its center, its rotation curve flattens at 1-2 disk scale lengths, it has the same specific angular momentum as local disks, its properties are consistent with the local K band TFR. Although mainly rotationally supported, its gas component is dynamically heated with respect to local galaxies (V/sigma_z ~ 4) and it is offset from the local stellar mass TFR at the 2sigma level. But, this offset depends on the SED modeling parameters. In particular, for a 2-component star formation history (SFH), F257 is in agreement with the local stellar mass TFR. F257 is then a nearly (~75%) maximum disk. The dynamical properties of F257 are more like those of local galaxies than those of any other galaxy at similar redshift observed to date. However, the gas-to-stellar mass ratio is unusally large: 2.5.

We present the evolution of the luminosity-size and stellar mass-size relations of luminous (L_V>3.4x10^10h_70^-2L_sun) and of massive (M_*>3x10^10h_70^-2M_sun) galaxies in the last ~11 Gyr. We use very deep near-infrared images of the Hubble Deep Field-South and the MS1054-03 field in the J_s, H and K_s bands from FIRES to retrieve the sizes in the optical rest-frame for galaxies with z>1. We combine our results with those from GEMS at 0.2<z<1 and SDSS at z~0.1 to achieve a comprehensive picture of the optical rest-frame size evolution from z=0 to z=3. Galaxies are differentiated according to their light concentration using the Sersic index n. For less concentrated objects, the galaxies at a given luminosity were typically ~3+-0.5 (+-2 sigma) times smaller at z~2.5 than those we see today. The stellar mass-size relation has evolved less: the mean size at a given stellar mass was \~2+-0.5 times smaller at z~2.5, evolving proportional to (1+z)^{-0.40+-0.06}. Simple scaling relations between dark matter halos and baryons in a hierarchical cosmogony predict a stronger (although consistent within the error bars) than observed evolution of the stellar mass-size relation. The observed luminosity-size evolution out to z~2.5 matches well recent infall model predictions for Milky-Way type objects. For low-n galaxies, the evolution of the stellar mass-size relation would follow naturally if the individual galaxies grow inside-out. For highly concentrated objects, the situation is as follows: at a given luminosity, these galaxies were ~2.7+-1.1 times smaller at z~2.5 (or put differently, were typically ~2.2+-0.7 mag brighter at a given size than they are today), and at a given stellar mass the size has evolved proportional to (1+z)^{-0.45+-0.10}.

The evolution of the line width - luminosity relation for spiral galaxies, the Tully-Fisher relation, strongly constrains galaxy formation and evolution models. At this moment, the kinematics of z>1 spiral galaxies can only be measured using rest frame optical emission lines associated with star formation, such as Halpha and [OIII]5007/4959 and [OII]3727. This method has intrinsic difficulties and uncertainties. Moreover, observations of these lines are challenging for present day telescopes and techniques. Here, we present an overview of the intrinsic and observational challenges and some ways way to circumvent them. We illustrate our results with the HST/NICMOS grism sample data of z ~ 1.5 starburst galaxies. The number of galaxies we can use in the final Tully-Fisher analysis is only three. We find a ~2 mag offset from the local rest frame B and R band Tully-Fisher relation for this sample. This offset is partially explained by sample selection effects and sample specifics. Uncertainties in inclination and extinction and the effects of star formation on the luminosity can be accounted for. The largest remaining uncertainty is the line width / rotation curve velocity measurement. We show that high resolution, excellent seeing integral field spectroscopy will improve the situation. However, we note that no flat rotation curves have been observed for galaxies with z>1. This could be due to the described instrumental and observational limitations, but it might also mean that galaxies at z>1 have not reached the organised motions of the present day.

We investigate the nature of the substantial population of high-z galaxies with Js-Ks>2.3 discovered as part of our FIRES survey. This colour cut efficiently isolates z>2 galaxies with red rest-frame optical colors ("Distant Red Galaxies" or DRGs). We select objects in the 2.5'x2.5' HDF-South (HDF-S) and 5'x5' field around the MS1054-03 cluster; the surface densities at Ks<21 are 1.6+-0.6 and 1.0+-0.2 arcmin^-2. We discuss the 34 DRGs at 2<z<3.5: 11 at Ks<22.5 in HDF-S and 23 at Ks<21.7 in the MS1054-03 field. We analyze the SEDs constructed from our deep near-infrared (NIR) and optical imaging from the ESO VLT and HST. We develop diagnostics involving I-Js, Js-H, and H-Ks to argue that the red NIR colors of DRGs cannot be attributed solely to extinction and require for many an evolved stellar population with prominent Balmer/4000A break. In the rest-frame, the optical colours of DRGs fall within the envelope of normal nearby galaxies and the UV colours suggest a wide range in star formation activity and/or extinction. This contrasts with the much bluer and more uniform SEDs of Lyman break galaxies (LBGs). From evolutionary synthesis models with constant star formation, solar metallicity, Salpeter IMF, and Calzetti et al. extinction law, we derive for the HDF-S (MS1054-03 field) DRGs median ages of 1.7(2.0) Gyr, A_V = 2.7(2.4) mag, stellar masses 0.8(1.6)x10^11 Msun, M/L_V = 1.2(2.3) Msun/LVsun, and SFR = 120(170) Msun/yr. Models assuming declining SFRs with e-folding timescales of 10Myr-1Gyr generally imply younger ages, lower A_V's and SFRs, but similar stellar masses within a factor of two. Compared to LBGs at similar redshifts and rest-frame L_V's, DRGs are older, more massive, and more obscured for any given star formation history. [ABRIDGED]

We recently identified a substantial population of galaxies at z>2 with red rest-frame optical colors. These distant red galaxies (DRGs) are efficiently selected by the simple observed color criterion J-K>2.3. In this paper we present NIR spectroscopy with Keck/NIRSPEC of six DRGs at 2.4<z<3.2. We detect continuum emission and emission lines of all observed galaxies. Equivalent widths of H alpha are 20-30 Ang, smaller than measured for LBGs and nearby LIRGs, and comparable to normal nearby galaxies. The modest equivalent widths imply that the galaxies either have a decreasing star formation rate, or that they are very dusty. Fitting both the photometry and the H alpha lines, we find continuum extinction A_V=1-2 mag, ages 1-2.5 Gyr, star formation rates 200-400 solar masses/yr, and stellar masses 1-5x10^11 solar masses for models with constant star formation rates. From [NII]/H alpha ratios we infer that the metallicities are high, 1-1.5 x Solar. For four galaxies we can determine line widths from the optical emission lines. The widths are high, ranging from 130-240 km/s, and by combining data for LBGs and DRGs we find significant correlations between linewidth and restframe U-V color, and between linewidth and stellar mass. The latter correlation has a similar slope and offset as the ``baryonic Tully-Fisher relation'' for nearby galaxies. The median dynamical mass is ~2x10^11 solar masses, supporting the high stellar masses inferred from the photometry. We find that the median M/L_V ~ 0.8, a factor of ~5 higher than measured for LBGs. We infer from our small sample that DRGs are dustier, more metal rich, more massive, and have higher ages than z=3 LBGs of the same rest-frame V-band luminosity. Their high M/L ratios imply that they contribute significantly to the stellar mass density at z~2.5. [ABRIDGED]

The luminosity-size and stellar mass-size distributions of galaxies out to z~3 is presented. We use very deep near-infrared images of the Hubble Deep Field South in the J_s, H, and K_s bands, taken as part FIRES at the VLT, to follow the evolution of the optical rest-frame sizes of galaxies. For a total of 168 galaxies with K_{s,AB}<23.5, we find that the rest-frame V-band sizes r_{e,V} of luminous galaxies (<L_V>~2x10^{10}h^{-2}L_sun) at 2<z<3 are 3 times smaller than for equally luminous galaxies today. In contrast, the mass-size relation has evolved relatively little: the size at mass <M_*>~2x10^{10}h^{-2}M_sun, has changed by 20(+-20)% since z~2.5. Both results can be reconciled by the fact that the stellar M/L ratio is lower in the luminous high z galaxies than in nearby ones because they have young stellar populations. The lower incidence of large galaxies at z~3 seems to reflect the rarity of galaxies with high stellar mass.

We present the evolution of the rest-frame optical luminosity density, of the integrated rest-frame optical color, and of the stellar mass density for a sample of Ks-band selected galaxies in the HDF-S. We derived the luminosity density in the rest-frame U, B, and V-bands and found that the luminosity density increases by a factor of 1.9+-0.4, 2.9+-0.6, and 4.9+-1.0 in the V, B, and U rest-frame bands respectively between a redshift of 0.1 and 3.2. We derived the luminosity weighted mean cosmic (U-B)_rest and (B-V)_rest colors as a function of redshift. The colors bluen almost monotonically with increasing redshift; at z=0.1, the (U-B)_rest and (B-V)_rest colors are 0.16 and 0.75 respectively, while at z=2.8 they are -0.39 and 0.29 respectively. We derived the luminosity weighted mean M/LV using the correlation between (U-V)_rest and log_{10} M/LV which exists for a range in smooth SFHs and moderate extinctions. We have shown that the mean of individual M/LV estimates can overpredict the true value by ~70% while our method overpredicts the true values by only ~35%. We find that the universe at z~3 had ~10 times lower stellar mass density than it does today in galaxies with LV>1.4 \times 10^{10} h_{70}^-2 Lsol. 50% of the stellar mass of the universe was formed by $z~1-1.5. The rate of increase in the stellar mass density with decreasing redshift is similar to but above that for independent estimates from the HDF-N, but is slightly less than that predicted by the integral of the SFR(z) curve.

Using deep near-infrared imaging of the Hubble Deep Field South with ISAAC on the Very Large Telescope we find 6 large disk-like galaxies at redshifts z = 1.4-3.0. The galaxies, selected in K_s (2.2 micron), are regular and surprisingly large in the near-infrared (rest-frame optical), with face-on effective radii r_e = 0.65"-0.9" or 5.0-7.5 h_70^-1 kpc in a Lambda-CDM cosmology, comparable to the Milky Way. The surface brightness profiles are consistent with an exponential law over 2-3 effective radii. The WFPC2 morphologies in Hubble Space Telescope imaging (rest-frame UV) are irregular and show complex aggregates of star-forming regions ~2" (~15 h_70^-1 kpc) across, symmetrically distributed around the K_s-band centers. The spectral energy distributions show clear breaks in the rest-frame optical. The breaks are strongest in the central regions of the galaxies, and can be identified as the age-sensitive Balmer/4000 Angstrom break. The most straightforward interpretation is that these galaxies are large disk galaxies; deep NIR data are indispensable for this classification. The candidate disks constitute 50% of galaxies with L_V > 6 x 10^10 h_70^-2 L_sun at z = 1.4-3.0. This discovery was not expected on the basis of previously studied samples. In particular, the Hubble Deep Field North is deficient in large galaxies with the morphologies and profiles we report here.

We confirm spectroscopically the existence of a population of galaxies at z>~2 with rest-frame optical colors similar to normal nearby galaxies. The galaxies were identified by their red near-infrared colors in deep images obtained with ISAAC on the Very Large Telescope. Redshifts of six galaxies with Js-Ks>2.3 were measured from optical spectra obtained with the Keck Telescope. Five out of six are in the range 2.43<=z<=3.52, demonstrating that the Js-Ks color selection is quite efficient. The rest-frame ultraviolet spectra of confirmed z>2 galaxies display a range of properties, with two galaxies showing emission lines characteristic of AGN, two having Ly-alpha in emission, and one showing interstellar absorption lines only. Their full spectral energy distributions are well described by constant star formation models with ages 1.4-2.6 Gyr, except for one galaxy whose colors indicate a dusty starburst. The confirmed z>2 galaxies are very luminous, with Ks=19.2-19.9. Assuming that our bright spectroscopic sample is representative for the general population of Js-Ks selected objects, we find that the surface density of red z>~2 galaxies is ~0.9/arcmin^2 to Ks=21. The surface density is comparable to that of Lyman-break selected galaxies with Ks<21, when corrections are made for the different redshift distributions of the two samples. Although there will be some overlap between the two populations, most 'optical-break' galaxies are too faint in the rest-frame ultraviolet to be selected as Lyman-break galaxies. The most straightforward interpretation is that star formation in typical optical-break galaxies started earlier than in typical Lyman-break galaxies. Optical-break galaxies may be the oldest and most massive galaxies yet identified at z>2, and could evolve into early-type galaxies and bulges.

We use very deep near-infrared photometry of the Hubble Deep Field South taken with ISAAC/VLT to identify a population of high redshift galaxies with rest- frame optical colors similar to those of nearby galaxies. The galaxies are chosen by their infrared colors Js-Ks > 2.3, aimed at selecting galaxies with redshifts above 2. When applied to our dataset, we find 14 galaxies with Ks < 22.5, corresponding to a surface density of 3+-0.8 /arcmin**2. The photometric redshifts all lie above 1.9, with a median of 2.6 and a rms of 0.7. The spectral energy distributions of these galaxies show a wide range: one is very blue in the rest-frame UV, and satisfies the normal Lyman-break criteria for high redshift, star-forming galaxies. Others are quite red throughout the observed spectral range, and are extremely faint in the optical, with a median V = 26.6. Hence these galaxies would not be included in photometric samples based on optical ground-based data, and spectroscopic follow-up is difficult. The spectral energy distributions often show a prominent break, identified as the Balmer break or 4000 Ang. break. The median age is 1 Gyr when fit with a constant star formation model with dust, or 0.7 Gyr when fit with a single burst model. Although significantly younger ages cannot be excluded when a larger range of models is allowed, the results indicate that these galaxies are among the oldest at these redshifts. The volume density to Ks=22.5 is half that of Lyman-break galaxies at z = 3. Since the mass-to-light ratios of the red galaxies are likely to be higher, the stellar mass density is inferred to be comparable to that of Lyman-break galaxies. These red galaxies may be the descendants of galaxies which started to form stars at very high redshifts, and they may evolve into the most massive galaxies at low redshift.

We present deep near-infrared (NIR) Js, H, and Ks-band ISAAC imaging of the WFPC2 field of the HDF-S. The 2.5'x 2.5' high Galactic latitude field was observed with the VLT under the best seeing conditions with integration times amounting to 33.6 hours in Js, 32.3 hours in H, and 35.6 hours in Ks. We reach total AB magnitudes for point sources of 26.8, 26.2, and 26.2 respectively (3 sigma), which make it the deepest ground-based NIR observations to date, and the deepest Ks-band data in any field. The effective seeing of the coadded images is ~0.45" in Js, ~0.48" in H, and ~0.46" in Ks. Using published WFPC2 optical data, we constructed a Ks-limited multicolor catalog containing 833 sources down to Ks,tot ~< 26 (AB), of which 624 have seven-band optical-to-NIR photometry. These data allow us to select normal galaxies from their rest-frame optical properties to high redshift (z ~< 4). The observations, data reduction and properties of the final images are discussed, and we address the detection and photometry procedures that were used in making the catalog. In addition, we present deep number counts, color distributions and photometric redshifts of the HDF-S galaxies. We find that our faint Ks-band number counts are flatter than published counts in other deep fields, which might reflect cosmic variations or different analysis techniques. Compared to the HDF-N, we find many galaxies with very red V-H colors at photometric redshifts 1.95 < z < 3.5. These galaxies are bright in Ks with infrared colors redder than Js-Ks > 2.3 (in Johnson magnitudes). Because they are extremely faint in the observed optical, they would be missed by ultraviolet-optical selection techniques, such as the U-dropout method.

We present the first results from the Faint Infra-Red Extragalactic Survey (FIRES) of the Hubble Deep Field South. Using a combination of deep near infrared data obtained with ISAAC at the VLT with the WFPC2 HST data, we construct a K-band selected sample of 136 galaxies with K(AB)<=23.5. We derive zphot's using a newly developed technique which models the observed spectral energy distribution with a linear combination of empirical galaxy templates. Testing our zphot technique against spectroscopic redshifts in the HDF-N yields dz/(1+z)~0.07 for z<6. We show that we can derive realistic error estimates in zphot by combining the systematic uncertainties derived from the HDF-N with errors in zphot which depend on the observed flux errors. The redshift histogram of galaxies in the HDF-S shows distinct structure with a sharp peak at z~0.5 and a broad enhancement at z~1-1.4. We find that 12% of our galaxies with K(vega)<21 lie at z>=2. While this is higher than the fraction predicted in Omega_M=1 hierarchical models of galaxy formation we find that published predictions using pure luminosity evolution models produce too many bright galaxies at redshifts greater than unity. Finally, we use our broad wavelength coverage to measure the rest-frame UBV luminosities Lrest for z<=3. There is a paucity of galaxies brighter than Lrest_V>=1.4e10 h^{-2}Lsun at z~1.5-2, however, at z>2 we find very luminous galaxies with Lrest_V>=5e10 h^{-2}Lsun (for Omega_M=0.3,Lambda=0.7, H_o=100 h km s^{-1}Mpc^{-1}). Local B-band luminosity functions predict 0.1 galaxies in the redshift range 2<=z<=3.5 and with Lrest_B>= 5e10 h^{-2}Lsun,B but we find 9. The discrepancy can be explained if L*_B increases by a factor of 2.4-3.2 with respect to locally determined values (abridged).