Strongly baryon-dominated disk galaxies at the peak of galaxy formation ten billion years ago

Journal name:
Nature
Volume:
543,
Pages:
397–401
Date published:
DOI:
doi:10.1038/nature21685
Received
Accepted
Published online

In the cold dark matter cosmology, the baryonic components of galaxies—stars and gas—are thought to be mixed with and embedded in non-baryonic and non-relativistic dark matter, which dominates the total mass of the galaxy and its dark-matter halo1. In the local (low-redshift) Universe, the mass of dark matter within a galactic disk increases with disk radius, becoming appreciable and then dominant in the outer, baryonic regions of the disks of star-forming galaxies. This results in rotation velocities of the visible matter within the disk that are constant or increasing with disk radius—a hallmark of the dark-matter model2. Comparisons between the dynamical mass, inferred from these velocities in rotational equilibrium, and the sum of the stellar and cold-gas mass at the peak epoch of galaxy formation ten billion years ago, inferred from ancillary data, suggest high baryon fractions in the inner, star-forming regions of the disks3, 4, 5, 6. Although this implied baryon fraction may be larger than in the local Universe, the systematic uncertainties (owing to the chosen stellar initial-mass function and the calibration of gas masses) render such comparisons inconclusive in terms of the mass of dark matter7. Here we report rotation curves (showing rotation velocity as a function of disk radius) for the outer disks of six massive star-forming galaxies, and find that the rotation velocities are not constant, but decrease with radius. We propose that this trend arises because of a combination of two main factors: first, a large fraction of the massive high-redshift galaxy population was strongly baryon-dominated, with dark matter playing a smaller part than in the local Universe; and second, the large velocity dispersion in high-redshift disks introduces a substantial pressure term that leads to a decrease in rotation velocity with increasing radius. The effect of both factors appears to increase with redshift. Qualitatively, the observations suggest that baryons in the early (high-redshift) Universe efficiently condensed at the centres of dark-matter haloes when gas fractions were high and dark matter was less concentrated.

At a glance

Figures

  1. Hα gas dynamics from KMOS and SINFONI in six massive star-forming galaxies.
    Figure 1: Hα gas dynamics from KMOS and SINFONI in six massive star-forming galaxies.

    The galaxies have redshifts between z = 0.9 and z = 2.4. KMOS provides seeing-limited data (full-width at half-maximum, FWHM ≈ 0.6″); SINFONI allows both seeing-limited and adaptive-optics-assisted observations (FWHM ≈ 0.2″). a, For each galaxy, the distribution of the integrated Hα line surface brightness is shown (colour scale; with a linear scaling), superposed on the stellar surface density (Σ*; top three panels) or the H-band continuum surface brightness (ΣH; bottom three panels) (white contours; square-root scaling). The horizontal white bar denotes the physical size scale. b, Velocity maps (colour scale; with extreme values indicated, in km s−1) superposed on Σ* or ΣH contours (white lines; square-root scaling), derived from fitting a Gaussian line profile to the Hα data in each pixel (0.05″). All galaxies have FWHM ≈ 0.25″ (2 kpc) except GS4 43501 and COS4 01351, which have FWHM ≈ 0.55″-0.67″ (5 kpc). c, Extracted line centroids (and ±1 r.m.s. uncertainties) along the kinematic major axis (white line in b). For COS4 01351 (bottom panel) and GS4 43501 (fourth panel), we show SINFONI (black filled circles) and KMOS (open blue circles) datasets; for D3a 15504 (top panel) we show SINFONI datasets at 0.2″ (filled black circles) and 0.5″ (open blue circles) resolution. Red continuous lines denote the best-fit dynamical model, constructed from a combination of a central compact bulge, an exponential disk and an NFW halo without adiabatic contraction, with a concentration of c = 4 at z ≈ 2 and c = 6.5 at z ≈ 1. For the modelling of the disk rotation, we also take into account the asymmetric drift correction inferred from the velocity dispersion curves (d and e; ref. 14). The times shown in each panel indicate the total on-source integration time. d, e, Two-dimensional (d; colour scale) and major-axis (e; with ±1 r.m.s. uncertainties) velocity dispersion distributions inferred from the Gaussian fits (after removal of the instrumental response: σinstr ≈ 37 km s−1 at z ≈ 0.85 and z ≈ 2.2; σinstr ≈ 45 km s−1 at z = 1.5), superposed on Σ* or ΣH contours (d; white lines). The numbers in d indicate the minimum and maximum velocity dispersions. The colouring of the data and red lines in e are as in c. All physical units are based on a concordance, flat cold dark matter cosmology, with cosmological constant Λ, matter density relative to the critical density of closing the Universe Ωm = 0.3 and ratio of baryonic to total matter density Ωbaryon/Ωm = 0.17, and for the z = 0 Hubble parameter H0 = 70 km s−1 Mpc−1.

  2. Normalized rotation curves.
    Figure 2: Normalized rotation curves.

    a, The various symbols denote the folded and binned rotation curve data for the six galaxies in Fig. 1, combined with the stacked rotation curve of 97 z = 0.6–2.6 star-forming galaxies18 (Methods). For all rotation curves, we averaged data points located symmetrically on either side of the dynamical centres, and plot the rotation velocities normalized by the maximum rotation velocity against the radii R normalized by the radius at which the amplitude of the rotation velocity is maximum (|vrot| = vmax), Rmax. Error bars are ±1 r.m.s. b, The black data denote the binned averages of the six individual galaxies, as well as the stack shown in a, with 1 r.m.s. uncertainties of the error-weighted means shown as grey shading (the outermost point has lighter shading to indicate that only two data points entered the average). In individual galaxies, Rmax depends on the ratio of bulge to total baryonic mass of the galaxy, the size of the galaxy and the instrumental resolution, leading to varying amounts of beam smearing. We assume an average value of Rmax ≈ (1.3–1.5)R1/2. For comparison, the grey dashed line indicates the slope of a typical rotation curve for low-redshift (z = 0), massive (log(M*/M) ≈ 11), star-forming disk galaxies19, which are comparable to the six galaxies we studied; the dotted red and solid green curves are the rotation curves of M31 (the Andromeda galaxy)17 and the Milky Way20. The thick magenta curve is the rotation curve of an infinitely thin, ‘Freeman’ exponential disk21 with Sérsic index nSersic = 1. The blue curve is a turbulent, thick exponential disk, including ‘asymmetric drift’ corrections for an assumed radially constant velocity dispersion of σ0 ≈ 50 km s−1 (and a ratio of rotation velocity to dispersion of vrot/σ0 ≈ 5)17.

  3. Dark-matter fractions.
    Figure 3: Dark-matter fractions.

    Dark-matter fractions fDM from different methods are listed as a function of the circular velocity of the disk vc, at approximately the half-light radius of the disk R1/2, for galaxies in the current Universe and about 10 Gyr ago. The large blue circles with red outlines indicate the dark-matter fractions derived from the outer-disk rotation curves for the six high-redshift disks presented here (Table 1), along with the ±2 r.m.s. uncertainties of the inferred dark-matter fractions and circular velocities. The average dark-matter fractions (and their ±1 r.m.s. errors/dispersions in the two coordinates) obtained from the comparison of inner rotation curves and the sum of stellar and gas masses for 92 z = 2–2.6 and 106 z = 0.6–1.1 star-forming galaxies are indicated by the filled black circle and green triangle, respectively6.We compare these high-redshift results to z = 0 estimates obtained using different independent techniques for late-type, star-forming disks7 (crossed grey squares, red filled square), for the Milky Way20 (crossed black circle), for massive, bulged, lensed disks7 (crossed red circles), and for passive early-type disks23 (thick magenta line). The upward magenta arrow marks the typical change if the z = 0 data dark-matter haloes are maximally adiabatically contracted.

  4. Location of the galaxies included in our analysis.
    Extended Data Fig. 1: Location of the galaxies included in our analysis.

    a, Location in stellar-mass–star-formation-rate space. The star-formation rate (SFR) is normalized to that of the ‘main sequence’37 at the redshift and stellar mass of each galaxy . b, Location in stellar-mass–size space. The size is the half-light radius measured in the observed H-band corrected to the rest-frame 5,000 Å and normalized to that of the mass–size relation for star-forming galaxies26 at the redshift and stellar mass of each source . In a and b, the greyscale image shows the distribution of the underlying galaxy population at 0.7 < z < 2.7 taken from the 3D-HST source catalogue at log(M*/M) > 9.0 and KAB < 23 mag (the magnitude cut applied when selecting KMOS3D targets and corresponding roughly to the completeness limits of the parent samples for SINS/zC-SINF targets). The current 2.5-year KMOS3D sample is shown with blue circles, and the SINS/zC-SINF sample with green diamonds. The two KMOS3D and four SINS/zC-SINF galaxies with individual outer rotation curves (RCs) are plotted as yellow circles and diamonds, respectively. Similarly, the KMOS3D and SINS/zC-SINF galaxies included in the stacked rotation curve are plotted as red circles and diamonds. All 3D-HST and KMOS3D galaxies are included in a, whereas only star-forming galaxies (SFGs) are shown in b, defined as having a specific star-formation rate higher than the inverse of the Hubble time at their redshift. The galaxies with individual outer rotation curves lie on and up to a factor of four times the main-sequence (MS) in star-formation rate (with mean and median log(SFR/SFRMS) = 0.24), and have sizes 1.2–2 times the relation (‘M–R SFGs’; mean and median offset in ). In star-formation rate and , the distribution of the stacked rotation-curve sample is essentially the same as the reference 3D-HST population in mean/median offsets (approximately 0.06 dex above the main-sequence and 0.07 dex above the mass–size relation) and in their scatter about the relationships (approximately 0.3 dex in log(SFR) and 0.17 dex in ); see refs 26, 37.

  5. Quality of fit and error of parameter determinations.
    Extended Data Fig. 2: Quality of fit and error of parameter determinations.

    The reduced chi-squared as a function of the dark-matter fraction fDM at R1/2 for the six galaxies in our sample, once the other parameters (x0, y0, the position angle of the kinematic major axis PAmaj, i, σ0, R1/2 and B/T) are fixed at their best-fit values. Global minima are marked by circles; error bars give Δχ2 = ±4 ranges, corresponding to confidence levels of 95% (2 r.m.s.) under the assumption of single-parameter Gaussian distributions. This is the most important parameter dependence for our dataset.

  6. Mean changes in fDM and  for changes in the secondary parameters B/T and R1/2, for COS 01351, D3a 6397, GS4 43501 and D3a 15504.
    Extended Data Fig. 3: Mean changes in fDM and for changes in the secondary parameters B/T and R1/2, for COS 01351, D3a 6397, GS4 43501 and D3a 15504.

    Changes in B/T and R1/2 are labelled ‘B/T ± 0.1’ and ‘Re ± 1σ’, respectively, where 1σ is the uncertainty on R1/2 given in Table 1; is the reduced chi-squared.

  7. Cumulative mass as a function of radius for one of our studied galaxies (GS4 43501).
    Extended Data Fig. 4: Cumulative mass as a function of radius for one of our studied galaxies (GS4 43501).

    Solid lines show the best fit; error bars show the variations in total (black, grey), baryonic (green) and dark-matter (DM; purple) mass at the outermost projected radius constrained by our data, if deviations from B/T and R1/2 within the uncertainties are considered (only cases with are considered). Dashed lines show the best fit for a model with lower concentration parameter (c = 2 instead of c = 5); dashed-dotted lines show the best fit for a model with adiabatic contraction (AC)97. Both modifications of the dark-matter profile lead to changes in the cumulative mass that are smaller than those obtained by varying B/T and R1/2 within the above uncertainties. The grey lines encompass variations in the dark-matter fraction of fDM(R1/2) = [0.14, 0.27] (best-fit fDM(R1/2) = 0.19).

  8. Minor axis cut at Rmajor = 0.71″ of D3a 15504.
    Extended Data Fig. 5: Minor axis cut at Rmajor = 0.71″ of D3a 15504.

    Shown are the velocities (data points, with 1 r.m.s. error bars) and disk models for different inclinations (lines): 25° (red), 30° (blue), 40° (magenta) and 50° (green). The minor-axis cut favours a low inclination. In combination with the morphology of the stellar surface density distribution (Fig. 1) and the constraint on the baryonic mass of the disk, this yields an overall inclination of 34° ± 5° (Table 1). Rmajor is the radial distance from the centre of the galaxy along the kinematic major axis.

  9. Residual maps.
    Extended Data Fig. 6: Residual maps.

    a, b, Residual maps (data minus model) for velocity (a; vdata − vmodel) and velocity dispersion (b; σdata − σmodel), for the six galaxies studied here. The colour scale is the same in all maps (from −200 km s−1 (purple) to +200 km s−1 (white)). Minimum and maximum values are noted in each map, as are the median and median dispersion (‘disp’) values.

References

  1. White, S. D. M. & Rees, M. J. Core condensation in heavy halos: a two-stage theory for galaxy formation and clustering. Mon. Not. R. Astron. Soc. 183, 341358 (1978)
  2. Sofue, Y. & Rubin, V. Rotation curves of spiral galaxies. Annu. Rev. Astron. Astrophys. 39, 137174 (2001)
  3. Förster Schreiber, N. M. et al. The SINS survey: SINFONI integral field spectroscopy of z ~ 2 star-forming galaxies. Astrophys. J. 706, 13641428 (2009)
  4. van Dokkum, P. G. et al. Forming compact massive galaxies. Astrophys. J. 813, 23 (2015)
  5. Price, S. H. et al. The MOSDEF survey: dynamical and baryonic masses and kinematic structures of star-forming galaxies at 1.4 ≤ z ≤ 2.6. Astrophys. J. 819, 80 (2016)
  6. Wuyts, S. et al. KMOS3D: dynamical constraints on the mass budget in early star-forming disks. Astrophys. J. 831, 149 (2016)
  7. Courteau, S. & Dutton, A. A. On the global mass distribution in disk galaxies. Astrophys. J. 801, L20 (2015)
  8. Wisnioski, E. et al. The KMOS3D survey: design, first results, and the evolution of galaxy kinematics from 0.7 ≤ z ≤ 2.7. Astrophys. J. 799, 209 (2015)
  9. Grogin, N. A. et al. CANDELS: the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey. Astrophys. J. Suppl. Ser. 197, 35 (2011)
  10. Koekemoer, A. M. et al. CANDELS: the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey—the Hubble Space Telescope observations, imaging data products, and mosaics. Astrophys. J. Suppl. Ser. 197, 36 (2011)
  11. Kong, X. et al. A wide area survey for high-redshift massive galaxies. I. Number counts and clustering of BzKs and EROs. Astrophys. J. 638, 7287 (2006)
  12. Brammer, G. B. et al. 3D-HST: a wide-field grism spectroscopic survey with the Hubble Space Telescope. Astrophys. J. Suppl. Ser. 200, 13 (2012)
  13. Momcheva, I. G. et al. The 3D-HST survey: Hubble Space Telescope WFC3/G141 grism spectra, redshifts, and emission line measurements for ~100,000 galaxies. Astrophys. J. Suppl. Ser. 225, 27 (2016)
  14. Burkert, A. et al. High-redshift star-forming galaxies: angular momentum and baryon fraction, turbulent pressure effects, and the origin of turbulence. Astrophys. J. 725, 23242332 (2010)
  15. Casertano, S. & van Gorkom, J. H. Declining rotation curves: the end of a conspiracy? Astron. J. 101, 12311241 (1991)
  16. Honma, M. & Sofue, Y. On the Keplerian rotation curves of galaxies. Publ. Astron. Soc. Jap. 49, 539545 (1997)
  17. Carignan, C., Chemin, L., Huchtmeier, W. K. & Lockman, F. J. The extended H i rotation curve and mass distribution of M31. Astrophys. J. 641, L109L112 (2006)
  18. Lang, P. et al. Falling outer rotation curves of star-forming galaxies at0.6 ≤ z ≤2.6 probed with KMOS3D and SINS/zC-SINF (submitted)
  19. Catinella, B., Giovanelli, R. & Haynes, M. P. Template rotation curves for disk galaxies. Astrophys. J. 640, 751761 (2006)
  20. Bland-Hawthorn, J. & Gerhard, O. The galaxy in context: structural, kinematic, and integrated properties. Annu. Rev. Astron. Astrophys. 54, 529596 (2016)
  21. Freeman, K. On the disks of spiral and S0 galaxies. Astrophys. J. 160, 811830 (1970)
  22. Navarro, J. F., Frenk, C. S. & White, S. D. M. A universal density profile from hierarchical clustering. Astrophys. J. 490, 493508 (1997)
  23. Cappellari, M. et al. The ATLAS3D project – XV. Benchmark for early-type galaxies scaling relations from 260 dynamical models: mass-to-light ratio, dark matter, fundamental plane and mass plane. Mon. Not. R. Astron. Soc. 432, 17091741 (2013)
  24. Peng, Y. et al. Mass and environment as drivers of galaxy evolution in SDSS and zCOSMOS and the origin of the Schechter function. Astrophys. J. 721, 193221 (2010)
  25. Genzel, R. et al. Combined CO and dust scaling relations of depletion time and molecular gas fractions with cosmic time, specific star-formation rate, and stellar mass. Astrophys. J. 800, 20 (2015)
  26. van der Wel, A. et al. 3D-HST+CANDELS: the evolution of the galaxy size-mass distribution since z = 3. Astrophys. J. 788, 28 (2014)
  27. Guo, Q. & White, S. D. M. Galaxy growth in the concordance ΛCDM cosmology. Mon. Not. R. Astron. Soc. 384, 210 (2008)
  28. Oser, L., Ostriker, J. P., Naab, T., Johansson, P. H. & Burkert, A. The two phases of galaxy formation. Astrophys. J. 725, 23122323 (2010)
  29. Zolotov, A. et al. Compaction and quenching of high-z galaxies in cosmological simulations: blue and red nuggets. Mon. Not. R. Astron. Soc. 450, 23272353 (2015)
  30. Moster, B. P., Naab, T. & White, S. D. M. Galactic star formation and accretion histories from matching galaxies to dark matter haloes. Mon. Not. R. Astron. Soc. 428, 31213138 (2013)
  31. Mancini, C. et al. The zCOSMOS-SINFONI Project. I. Sample selection and natural-seeing observations. Astrophys. J. 743, 86 (2011)
  32. Eisenhauer, F. et al. The Universe in 3D: first observations with SPIFFI, the infrared integral field spectrometer for the VLT. The Messenger 113, 1725 (2003)
  33. Bonnet, H. et al. First light of SINFONI at the VLT. The Messenger 117, 1724 (2004)
  34. Sharples, R. M. et al. First light for the KMOS multi-object integral field spectrometer. The Messenger 151, 2123 (2012)
  35. Burkert, A. et al. The angular momentum distribution and baryon content of star-forming galaxies at z ~ 1-3. Astrophys. J. 826, 214 (2016)
  36. Skelton, R. E. et al. 3D-HST WFC4-selected photometric catalogs in the five CANDELS/3D-HST fields: photometry, photometric redshifts, and stellar masses. Astrophys. J. Suppl. Ser. 214, 24 (2014)
  37. Whitaker, K. E. et al. Constraining the low-mass slope of the star formation sequence at 0.5 < z < 2.5. Astrophys. J. 795, 104 (2014)
  38. van Dokkum, P. G. et al. First results from the 3D-HST survey: the striking diversity of massive galaxies at z > 1. Astrophys. J. 743, L15 (2011)
  39. Wuyts, S. et al. On star formation rates and star formation histories of galaxies out to z ~ 3. Astrophys. J. 738, 106 (2011)
  40. Bruzual, G. & Charlot, S. Stellar population synthesis at the resolution of 2003. Mon. Not. R. Astron. Soc. 344, 10001028 (2003)
  41. Calzetti, D. et al. The dust content and opacity of actively star-forming galaxies. Astrophys. J. 533, 682695 (2000)
  42. Chabrier, G. Galactic stellar and substellar initial mass function. Publ. Astron. Soc. Pac. 115, 763795 (2003)
  43. Wuyts, E. et al. A consistent study of metallicity evolution at 0.8 < z < 2.6. Astrophys. J. 789, L40 (2014)
  44. Wuyts, E. et al. The evolution of metallicity and metallicity gradients from z = 2.7 to 0.6 with KMOS3D. Astrophys. J. 827, 74 (2016)
  45. Erb, D. K. et al. The mass-metallicity relation at z ≥ 2. Astrophys. J. 644, 813828 (2006)
  46. Zahid, H. J., Kewley, L. J. & Bresoline, F. The mass-metallicity and luminosity-metallicity relations from DEEP2 at z ~ 0.8. Astrophys. J. 730, 137 (2011)
  47. Zahid, H. J. et al. The FMOS-COSMOS survey of star-forming galaxies at z ~ 1.6. II. The mass-metallicity relation and the dependence on star formation rate and dust extinction. Astrophys. J. 792, 75 (2014)
  48. Stott, J. P. et al. A fundamental metallicity relation for galaxies at z = 0.84–1.47 from HiZELS. Mon. Not. R. Astron. Soc. 436, 11301141 (2013)
  49. Steidel, C. C. et al. Strong nebular line ratios in the spectra of z ~ 2–3 star-forming galaxies: first results from KBSS-MOSFIRE. Astrophys. J. 795, 165 (2014)
  50. Sanders, R. L. et al. The MOSDEF survey: mass, metallicity, and star-formation rate at z ~ 2.3. Astrophys. J. 799, 138 (2015)
  51. Wuyts, S. et al. What do we learn from IRAC observations of galaxies at 2 < z < 3.5? Astrophys. J. 655, 5165 (2007)
  52. Kewley, L. J. et al. Theoretical evolution of optical strong lines across cosmic time. Astrophys. J. 774, 100 (2013)
  53. Tacconi, L. J. et al. PHIBSS: unified scaling relations of depletion time and molecular gas fractions with cosmic time, specific star formation rate and stellar mass. Preprint at https://arxiv.org/abs/1702.01140 (2017)
  54. Förster Schreiber, N. M. et al. Constraints on the assembly and dynamics of galaxies. I. Detailed rest-frame optical morphologies on kiloparsec scale of z ~ 2 star-forming galaxies. Astrophys. J. 731, 65 (2011)
  55. Förster Schreiber, N. M. et al. Constraints on the assembly and dynamics of galaxies. II. Properties of kiloparsec-scale clumps in rest-frame optical emission of z ~ 2 star-forming galaxies. Astrophys. J. 739, 45 (2011)
  56. Wuyts, S. et al. Smooth(er) stellar mass maps in CANDELS: constrains on the longevity of clumps in high-redshift star-forming galaxies. Astrophys. J. 753, 114 (2012)
  57. Lang, P. et al. Bulge growth and quenching since z = 2.5 in CANDELS/3D-HST. Astrophys. J. 788, 11 (2014)
  58. Tacchella, S. et al. The confinement of star-forming galaxies into a main sequence through episodes of gas compaction, depletion and replenishment. Mon. Not. R. Astron. Soc. 457, 27902813 (2016)
  59. Tacchella, S. et al. Evolution of density profiles in high-z galaxies: compaction and quenching inside-out. Mon. Not. R. Astron. Soc. 458, 242263 (2016)
  60. Wuyts, S. et al. A CANDELS-3D-HST synergy: resolved star formation patterns at 0.7 < z < 1.5. Astrophys. J. 779, 135 (2013)
  61. Genzel, R. et al. The SINS/zC-SINF survey of z ~ 2 galaxy kinematics: evidence for gravitational quenching. Astrophys. J. 785, 75 (2014)
  62. Nelson, E. J. et al. Where stars form: inside-out growth and coherent star formation from HST Hα maps of 3200 galaxies across the main sequence at 0.7 < z < 1.5. Astrophys. J. 828, 27 (2016)
  63. Tacchella, S. et al. Evidence for mature bulges and an inside-out quenching phase 3 billion years after the Big Bang. Science 348, 314317 (2015)
  64. Tacchella, S. et al. SINS/zC-SINF survey of z ~ 2 galaxy kinematics: rest-frame morphology, structure, and colors from near-infrared Hubble Space Telescope imaging. Astrophys. J. 802, 101 (2015)
  65. Wuyts, S. et al. Galaxy structure and mode of star formation in the SFR-mass plane from z ~ 2.5 to z ~ 0.1. Astrophys. J. 742, 96 (2011)
  66. Bell, E. F. et al. What turns galaxies off? The different morphologies of star-forming and quiescent galaxies since z ~ 2 from CANDELS. Astrophys. J. 753, 167 (2012)
  67. Bruce, V. A. et al. The bulge–disc decomposed evolution of massive galaxies at 1 < z < 3 in CANDELS. Mon. Not. R. Astron. Soc. 444, 10011033 (2014)
  68. Bruce, V. A. et al. The decomposed bulge and disc size–mass relations of massive galaxies at 1 < z < 3 in CANDELS. Mon. Not. R. Astron. Soc. 444, 16601673 (2014)
  69. Nelson, E. J. et al. Spatially resolved dust maps from Balmer decrements in galaxies at z ~ 1.4. Astrophys. J. 817, L9 (2016)
  70. Tadaki, K. et al. SXDF-ALMA 1.5 arcmin2 deep survey: a compact dusty star-forming galaxy at z = 2.5. Astrophys. J. 811, L3 (2015)
  71. Tadaki, K.-i. et al. Bulge-forming galaxies with an extended rotating disk at z ~ 2. Astrophys. J. 834, 135 (2017)
  72. Barro, G. et al. Sub-kiloparsec ALMA imaging of compact star-forming galaxies at z ~ 2.5: revealing the formation of dense galactic cores in the progenitors of compact quiescent galaxies. Astrophys. J. 827, L32 (2016)
  73. van der Wel, A. et al. Geometry of star-forming galaxies from SDSS, 3D-HST, and CANDELS. Astrophys. J. 792, L6 (2014)
  74. Elmegreen, D. M. et al. Galaxy morphologies in the Hubble ultra deep field: dominance of linear structures at the detection limit. Astrophys. J. 631, 85100 (2005)
  75. Law, D. R. et al. An HST/WFC3-IR morphological survey of galaxies at z = 1.5–3.6. I. Survey description and morphological properties of star-forming galaxies. Astrophys. J. 745, 85 (2012)
  76. Genzel, R. et al. The rapid formation of a large rotating disk galaxy three billion years after the Big Bang. Nature 442, 786789 (2006)
  77. Kassin, S. et al. The epoch of disk settling: z ~ 1 to now. Astrophys. J. 758, 106 (2012)
  78. Newman, S. F. et al. The SINS/zC-SINF survey of z ~ 2 galaxy kinematics: the nature of dispersion-dominated galaxies. Astrophys. J. 767, 104 (2013)
  79. Tacconi, L. J. et al. Phibss: molecular gas content and scaling relations in z ~ 1–3 massive, main-sequence star-forming galaxies. Astrophys. J. 768, 74 (2013)
  80. Davies, R. et al. How well can we measure the intrinsic velocity dispersion of distant disk galaxies? Astrophys. J. 741, 69 (2011)
  81. van der Kruit, P. C. & Allen, R. J. The kinematics of spiral and irregular galaxies. Annu. Rev. Astron. Astrophys. 16, 103139 (1978)
  82. Genzel, R. et al. The Sins survey of z ~ 2 galaxy kinematics: properties of the giant star-forming clumps. Astrophys. J. 733, 101 (2011)
  83. Newman, S. F. et al. Shocked superwinds from the z ~ 2 clumpy star-forming galaxy, ZC406690. Astrophys. J. 752, 111 (2012)
  84. Förster Schreiber, N. M. et al. The SINS/zC-SINF survey of z ~ 2 galaxy kinematics: evidence for powerful active galactic nucleus-driven nuclear outflows in massive star-forming galaxies. Astrophys. J. 787, 38 (2014)
  85. Cappellari, M. Structure and kinematics of early-type galaxies from integral field spectroscopy. Annu. Rev. Astron. Astrophys. 54, 597665 (2016)
  86. Noordermeer, E. The rotation curves of flattened Sérsic bulges. Mon. Not. R. Astron. Soc. 385, 13591364 (2008)
  87. Bullock, J. S. et al. Profiles of dark haloes: evolution, scatter and environment. Mon. Not. R. Astron. Soc. 321, 559575 (2001)
  88. Gao, L. et al. The redshift dependence of the structure of massive Λ cold dark matter haloes. Mon. Not. R. Astron. Soc. 387, 536544 (2008)
  89. Dutton, A. A. & Macciò, A. V. Cold dark matter haloes in the Planck era: evolution of structural parameters for Einasto and NFW profiles. Mon. Not. R. Astron. Soc. 441, 33593374 (2014)
  90. Johansson, P. H., Naab, T. & Ostriker, J. P. Gravitational heating helps make massive galaxies red and dead. Astrophys. J. 697, L38L43 (2009)
  91. Dutton, A. A. et al. The response of dark matter haloes to elliptical galaxy formation: a new test for quenching scenarios. Mon. Not. R. Astron. Soc. 453, 24472464 (2015)
  92. Duffy, A. R. et al. Impact of baryon physics on dark matter structures: a detailed simulation study of halo density profiles. Mon. Not. R. Astron. Soc. 405, 21612178 (2010)
  93. Pontzen, A. & Governato, F. How supernova feedback turns dark matter cusps into cores. Mon. Not. R. Astron. Soc. 421, 34643471 (2012)
  94. Martizzi, D., Teyssier, R. & Moore, B. Cusp–core transformations induced by AGN feedback in the progenitors of cluster galaxies. Mon. Not. R. Astron. Soc. 432, 19471954 (2013)
  95. Dutton, A. A. et al. NIHAO IX: the role of gas inflows and outflows in driving the contraction and expansion of cold dark matter haloes. Mon. Not. R. Astron. Soc. 461, 26582675 (2016)
  96. Lilly, S. J. et al. Gas regulation of galaxies: the evolution of the cosmic specific star formation rate, the metallicity-mass-star-formation rate relation, and the stellar content of halos. Astrophys. J. 772, 119 (2013)
  97. Blumenthal, G. R. et al. Contraction of dark matter galactic halos due to baryonic infall. Astrophys. J. 301, 2734 (1986)
  98. Binney, J. & Tremaine, S. Galactic Dynamics 2nd edn (Princeton Univ. Press, 2008)
  99. van der Kruit, P. C. & Freeman, K. C. Galaxy disks. Annu. Rev. Astron. Astrophys. 49, 301371 (2011)
  100. Toomre, A. On the gravitational stability of a disk of stars. Astrophys. J. 139, 12171238 (1964)
  101. Merritt, D. & Sellwood, J. A. Bending instabilities in stellar systems. Astrophys. J. 425, 551567 (1994)
  102. Levine, E. S., Blitz, L. & Heiles, C. The vertical structure of the outer Milky Way H i disk. Astrophys. J. 643, 881896 (2006)
  103. Courteau, S. et al. Galaxy masses. Rev. Mod. Phys. 86, 47119 (2014)
  104. Genel, S. et al. Introducing the Illustris project: the evolution of galaxy populations across cosmic time. Mon. Not. R. Astron. Soc. 445, 175200 (2014)
  105. Schaye, J. et al. The EAGLE project: simulating the evolution and assembly of galaxies and their environments. Mon. Not. R. Astron. Soc. 446, 521554 (2015)
  106. Johansson, P. H., Naab, T. & Ostriker, J. P. Forming early-type galaxies in ΛCDM simulations. I. Assembly histories. Astrophys. J. 754, 115 (2012)
  107. Anglés-Alcázar, D. et al. Cosmological zoom simulations of z = 2 galaxies: the impact of galactic outflows. Astrophys. J. 782, 84 (2014)
  108. Governato, F. et al. Forming disc galaxies in ΛCDM simulations. Mon. Not. R. Astron. Soc. 374, 14791494 (2007)
  109. Scannapieco, C. et al. The formation and survival of discs in a ΛCDM universe. Mon. Not. R. Astron. Soc. 396, 696708 (2009)
  110. Scannapieco, C. et al. The Aquila comparison project: the effects of feedback and numerical methods on simulations of galaxy formation. Mon. Not. R. Astron. Soc. 423, 17261749 (2012)
  111. Brook, C. B. et al. Hierarchical formation of bulgeless galaxies – II. Redistribution of angular momentum via galactic fountains. Mon. Not. R. Astron. Soc. 419, 771779 (2012)
  112. Agertz, O., Teyssier, R. & Moore, B. The formation of disc galaxies in a ΛCDM universe. Mon. Not. R. Astron. Soc. 410, 13911408 (2011)
  113. Aumer, M., White, S. D. M. & Naab, T. Towards a more realistic population of bright spiral galaxies in cosmological simulations. Mon. Not. R. Astron. Soc. 434, 31423164 (2013)
  114. Hopkins, P. F. et al. Galaxies on FIRE (Feedback In Realistic Environments): stellar feedback explains cosmologically inefficient star formation. Mon. Not. R. Astron. Soc. 445, 581603 (2014)
  115. Marinacci, F., Pakmor, R. & Springel, V. The formation of disc galaxies in high-resolution moving-mesh cosmological simulations. Mon. Not. R. Astron. Soc. 437, 17501775 (2014)
  116. Übler, H. et al. Why stellar feedback promotes disc formation in simulated galaxies. Mon. Not. R. Astron. Soc. 443, 20922111 (2014)
  117. Genel, S. et al. Galactic angular momentum in the Illustris simulations: feedback and the Hubble sequence. Astrophys. J. 804, L40 (2015)
  118. Fiacconi, D., Feldmann, R. & Mayer, L. The Argo simulation – II. The early build-up of the Hubble sequence. Mon. Not. R. Astron. Soc. 446, 19571972 (2015)
  119. Carollo, M. C. et al. Newly quenched galaxies as the cause for the apparent evolution in average size of the population. Astrophys. J. 773, 112 (2013)

Download references

Author information

Affiliations

  1. Max-Planck-Institut für extraterrestrische Physik (MPE), Giessenbachstrasse 1, 85748 Garching, Germany

    • R. Genzel,
    • N. M. Förster Schreiber,
    • H. Übler,
    • P. Lang,
    • R. Bender,
    • L. J. Tacconi,
    • E. Wisnioski,
    • S. Wuyts,
    • A. Beifiori,
    • S. Belli,
    • A. Burkert,
    • J. Chan,
    • R. Davies,
    • M. Fossati,
    • A. Galametz,
    • O. Gerhard,
    • D. Lutz,
    • J. T. Mendel,
    • E. J. Nelson,
    • R. Saglia,
    • K. Tadaki &
    • D. Wilman
  2. Departments of Physics and Astronomy, University of California, Berkeley, California 94720, USA

    • R. Genzel
  3. Max-Planck Institute for Astrophysics, Karl Schwarzschildstrasse 1, D-85748 Garching, Germany

    • T. Naab &
    • A. Burkert
  4. Universitäts-Sternwarte Ludwig-Maximilians-Universität (USM), Scheinerstrasse 1, D-81679 München, Germany

    • R. Bender,
    • A. Beifiori,
    • M. Fossati,
    • A. Galametz,
    • J. T. Mendel,
    • R. Saglia &
    • D. Wilman
  5. Department of Physics, University of Bath, Claverton Down, Bath BA2 7AY, UK

    • S. Wuyts
  6. Department of Particle Physics and Astrophysics, Faculty of Physics, The Weizmann Institute of Science, POB 26, Rehovot 76100, Israel

    • T. Alexander
  7. Space Telescope Science Institute, Baltimore, Maryland 21218, USA

    • G. Brammer
  8. Institute of Astronomy, Department of Physics, Eidgenössische Technische Hochschule, ETH Zürich, CH-8093 Zürich, Switzerland

    • C. M. Carollo &
    • S. Tacchella
  9. Center for Computational Astrophysics, 160 Fifth Avenue, New York, New York 10010, USA

    • S. Genel
  10. Department of Astronomy, Yale University, 260 Whitney Avenue, New Haven, Connecticut 06511, USA

    • I. Momcheva &
    • E. J. Nelson
  11. Osservatorio Astronomico di Padova, Vicolo dell’Osservatorio 5, Padova I-35122, Italy

    • A. Renzini
  12. School of Physics and Astronomy, Tel Aviv University, Tel Aviv 69978, Israel

    • A. Sternberg

Contributions

Drafting text, figures and Methods: R.G., N.M.F.S., H.Ü., T.N., L.J.T., O.G., D.L., A.R., R.S. and A.S.; data analysis and modelling: R.G., H.Ü., P.L., S.W. and R.D.; data acquisition and reduction: R.G., N.M.F.S., H.Ü., P.L., L.J.T., E.W., S.W., A.Be., S.B., J.C., M.F., A.G., J.T.M. and K.T.; KMOS3D and SINS/zC-SINF IFS survey design and management: N.M.F.S., R.B., E.W., C.M.C., A.R., R.S., S.T. and D.W.; 3D-HST survey analysis: N.M.F.S., S.W., G.B., I.M. and E.J.N.; theoretical interpretation: T.N., T.A., A.Bu., S.G. and O.G.

Competing financial interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to:

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Author details

Extended data figures and tables

Extended Data Figures

  1. Extended Data Figure 1: Location of the galaxies included in our analysis. (256 KB)

    a, Location in stellar-mass–star-formation-rate space. The star-formation rate (SFR) is normalized to that of the ‘main sequence’37 at the redshift and stellar mass of each galaxy . b, Location in stellar-mass–size space. The size is the half-light radius measured in the observed H-band corrected to the rest-frame 5,000 Å and normalized to that of the mass–size relation for star-forming galaxies26 at the redshift and stellar mass of each source . In a and b, the greyscale image shows the distribution of the underlying galaxy population at 0.7 < z < 2.7 taken from the 3D-HST source catalogue at log(M*/M) > 9.0 and KAB < 23 mag (the magnitude cut applied when selecting KMOS3D targets and corresponding roughly to the completeness limits of the parent samples for SINS/zC-SINF targets). The current 2.5-year KMOS3D sample is shown with blue circles, and the SINS/zC-SINF sample with green diamonds. The two KMOS3D and four SINS/zC-SINF galaxies with individual outer rotation curves (RCs) are plotted as yellow circles and diamonds, respectively. Similarly, the KMOS3D and SINS/zC-SINF galaxies included in the stacked rotation curve are plotted as red circles and diamonds. All 3D-HST and KMOS3D galaxies are included in a, whereas only star-forming galaxies (SFGs) are shown in b, defined as having a specific star-formation rate higher than the inverse of the Hubble time at their redshift. The galaxies with individual outer rotation curves lie on and up to a factor of four times the main-sequence (MS) in star-formation rate (with mean and median log(SFR/SFRMS) = 0.24), and have sizes 1.2–2 times the relation (‘M–R SFGs’; mean and median offset in ). In star-formation rate and , the distribution of the stacked rotation-curve sample is essentially the same as the reference 3D-HST population in mean/median offsets (approximately 0.06 dex above the main-sequence and 0.07 dex above the mass–size relation) and in their scatter about the relationships (approximately 0.3 dex in log(SFR) and 0.17 dex in ); see refs 26, 37.

  2. Extended Data Figure 2: Quality of fit and error of parameter determinations. (269 KB)

    The reduced chi-squared as a function of the dark-matter fraction fDM at R1/2 for the six galaxies in our sample, once the other parameters (x0, y0, the position angle of the kinematic major axis PAmaj, i, σ0, R1/2 and B/T) are fixed at their best-fit values. Global minima are marked by circles; error bars give Δχ2 = ±4 ranges, corresponding to confidence levels of 95% (2 r.m.s.) under the assumption of single-parameter Gaussian distributions. This is the most important parameter dependence for our dataset.

  3. Extended Data Figure 3: Mean changes in fDM and for changes in the secondary parameters B/T and R1/2, for COS 01351, D3a 6397, GS4 43501 and D3a 15504. (70 KB)

    Changes in B/T and R1/2 are labelled ‘B/T ± 0.1’ and ‘Re ± 1σ’, respectively, where 1σ is the uncertainty on R1/2 given in Table 1; is the reduced chi-squared.

  4. Extended Data Figure 4: Cumulative mass as a function of radius for one of our studied galaxies (GS4 43501). (185 KB)

    Solid lines show the best fit; error bars show the variations in total (black, grey), baryonic (green) and dark-matter (DM; purple) mass at the outermost projected radius constrained by our data, if deviations from B/T and R1/2 within the uncertainties are considered (only cases with are considered). Dashed lines show the best fit for a model with lower concentration parameter (c = 2 instead of c = 5); dashed-dotted lines show the best fit for a model with adiabatic contraction (AC)97. Both modifications of the dark-matter profile lead to changes in the cumulative mass that are smaller than those obtained by varying B/T and R1/2 within the above uncertainties. The grey lines encompass variations in the dark-matter fraction of fDM(R1/2) = [0.14, 0.27] (best-fit fDM(R1/2) = 0.19).

  5. Extended Data Figure 5: Minor axis cut at Rmajor = 0.71″ of D3a 15504. (292 KB)

    Shown are the velocities (data points, with 1 r.m.s. error bars) and disk models for different inclinations (lines): 25° (red), 30° (blue), 40° (magenta) and 50° (green). The minor-axis cut favours a low inclination. In combination with the morphology of the stellar surface density distribution (Fig. 1) and the constraint on the baryonic mass of the disk, this yields an overall inclination of 34° ± 5° (Table 1). Rmajor is the radial distance from the centre of the galaxy along the kinematic major axis.

  6. Extended Data Figure 6: Residual maps. (243 KB)

    a, b, Residual maps (data minus model) for velocity (a; vdata − vmodel) and velocity dispersion (b; σdata − σmodel), for the six galaxies studied here. The colour scale is the same in all maps (from −200 km s−1 (purple) to +200 km s−1 (white)). Minimum and maximum values are noted in each map, as are the median and median dispersion (‘disp’) values.

Supplementary information

PDF files

  1. Supplementary Discussion (145 KB)

    This file contains a comparison of the results to state-of-the-art cosmological simulations of galaxy formation, and a brief comment on the connection to the ‘thick disk’-phenomenon in local spiral galaxies.

Additional data