Search

We present a search for extreme emission line galaxies (EELGs) at z<1 in the COSMOS and North Ecliptic Pole (NEP) fields with imaging from Subaru/Hyper Suprime-Cam (HSC) and a combination of new and existing spectroscopy. We select EELGs on the basis of substantial excess flux in the z broadband, which is sensitive to H$\alpha$ at 0.3

Ultraluminous Lyman alpha (Lya) emitting galaxies (ULLAEs) with log L (Lya)>43.5 erg/s near the epoch of reionization (z>5) make up the bright end of the LAE luminosity function (LF) and may provide insight into the process of reionization, including the formation of ionized bubbles around these extreme systems. We present a spectroscopic LF for ULLAEs at z=5.7. We used data from the HEROES ~45 sq. deg Subaru Hyper Suprime-Cam survey, which is centered on the North Ecliptic Pole and has both broadband (grizY) and narrowband (NB816 and NB921) imaging, to select candidate ULLAEs based on a NB816 excess and a strong broadband Lyman break. We spectroscopically observed 17 ULLAE candidates with DEIMOS on Keck II. We confirmed 12 as LAEs at z=5.7, 9 of which are ULLAEs. The remaining sources are an AGN at z=5.7, an [OIII]5007 emitter at z=0.63, a red star, and two spectroscopic non-detections. Using the 9 confirmed ULLAEs, we construct a ULLAE LF at z=5.7. After applying a comprehensive incompleteness correction, we compare our new z=5.7 LF with our recent z=6.6 LF and with other LFs from the literature to look for evolution at the ultraluminous end. We find the overall ratio of the z=5.7 to z=6.6 ULLAE comoving number densities to be 1.92 (+1.12, -0.71), which corresponds to a LF offset of 0.28 (+0.20, -0.20) dex., Comment: Accepted for publication in ApJ. 13 pages, 8 Figures, 3 Tables. v2 integrates an erratum to correct the coordinates of NEP5.7AGN. This change has been made directly in the article text, tables, and figures, and does not affect the results or conclusions of the work

We present the luminosity function (LF) for ultraluminous Ly$\alpha$ emitting galaxies (LAEs) at z = 6.6. We define ultraluminous LAEs (ULLAEs) as galaxies with logL(Ly$\alpha$) > 43.5 erg s$^{-1}$. We select our main sample using the g', r', i', z', and NB921 observations of a wide-area (30 deg$^2$) Hyper Suprime-Cam survey of the North Ecliptic Pole (NEP) field. We select candidates with g', r', i' > 26, NB921 $\leq$ 23.5, and NB921 - z' $\leq$ 1.3. Using the DEIMOS spectrograph on Keck II, we confirm 9 of our 14 candidates as ULLAEs at z = 6.6 and the remaining 5 as an AGN at z = 6.6, two [OIII]$\lambda$5007 emitting galaxies at z = 0.84 and z = 0.85, and two non-detections. This emphasizes the need for full spectroscopic follow-up to determine accurate LFs. In constructing the ULLAE LF at z = 6.6, we combine our 9 NEP ULLAEs with two previously discovered and confirmed ULLAEs in the COSMOS field: CR7 and COLA1. We apply rigorous corrections for incompleteness based on simulations. We compare our ULLAE LF at z = 6.6 with LFs at z = 5.7 and z = 6.6 from the literature. Our data reject some previous LF normalizations and power law indices, but they are broadly consistent with others. Indeed, a comparative analysis of the different literature LFs suggests that none is fully consistent with any of the others, making it critical to determine the evolution from z = 5.7 to z = 6.6 using LFs constructed in exactly the same way at both redshifts., Comment: Accepted for publication in ApJ. 12 pages, 8 Figures, 3 Tables

We report on a search for ultraluminous Lyman alpha emitting galaxies (LAEs) at z=6.6 using the NB921 filter on Hyper Suprime-Cam on the Subaru telescope. We searched a 30 degree squared area around the North Ecliptic Pole, which we observed in broadband g', r', i', z', and y' and narrowband NB816 and NB921, for sources with NB921 < 23.5 and z' - NB921 > 1.3. This corresponds to a selection of log L(Ly-alpha) > 43.5 erg/s. We followed up seven candidate LAEs (out of thirteen) with the Keck DEIMOS spectrograph and confirmed five z=6.6 LAEs, one z=6.6 AGN with a broad Ly-alpha line and a strong red continuum, and one low-redshift ([OIII]5007) galaxy. The five ultraluminous LAEs have wider line profiles than lower luminosity LAEs, and one source, NEPLA4, has a complex line profile similar to that of COLA1. In combination with previous results, we show that the line profiles of the z=6.6 ultraluminous LAEs are systematically different than those of lower luminosity LAEs at this redshift. This result suggests that ultraluminous LAEs generate highly ionized regions of the intergalactic medium in their vicinity that allow the full Lyman alpha profile of the galaxy---including any blue wings---to be visible. If this interpretation is correct, then ultraluminous LAEs offer a unique opportunity to determine the properties of the ionized zones around them, which will help in understanding the ionization of the z ~ 7 intergalactic medium. A simple calculation gives a very rough estimate of 0.015 for the escape fraction of ionizing photons, but more sophisticated calculations are needed to fully characterize the uncertainties., Comment: Accepted for publication in ApJ. 13 pages, 13 figures

We examine the redshift evolution of standard strong emission-line diagnostics for Hbeta-selected star-forming galaxies using the local SDSS sample and a new z = 0.2 - 2.3 sample obtained from HST WFC3 grism and Keck DEIMOS and MOSFIRE data. We use the SDSS galaxies to show that there is a systematic dependence of the strong emission-line properties on Balmer-line luminosity, which we interpret as showing that both the N/O abundance and the ionization parameter increase with increasing line luminosity. Allowing for the luminosity dependence tightens the diagnostic diagrams and the metallicity calibrations. The combined SDSS and high-redshift samples show that there is no redshift evolution in the line properties once the luminosity correction is applied, i.e., all galaxies with a given L(Hbeta) have similar strong emission-line distributions at all the observed redshifts. We argue that the best metal diagnostic for the high-redshift galaxies may be a luminosity-adjusted version of the [NII]6584/Halpha metallicity relation., Comment: Accepted for publication in ApJ. 16 pages, 15 figures

We report the first space-based very deep near-infrared continuum observations of a uniform sample of z = 6.5 galaxies with log(L(L_alpha)) = 42.5 - 43 erg/s selected from narrow-band line searches and with spectroscopically confirmed Ly alpha emission. The 1.4 micron HST WFC3 observations are deep enough (AB(1 sigma) = 28.75) to measure individual continuum magnitudes at this redshift for all of the objects. We compare the results with continuum-selected samples at the same redshift and find that Ly alpha emission is present in 24% of all galaxies with M_AB(1350 A) < -20 at z = 6.5. The error in this quantity is dominated by systematic uncertainties, which could be as large as multiplicative factors of three. The Ly alpha galaxies are extended but small (size < 1 kpc), and have star formation rates of approximately 10 Msun/yr. We find a mean L(Ly alpha)/(nu L_nu) at 1400A of 0.08, with the seven objects showing a range from 0.026 to 0.26, implying that there is little sign of destruction of the Ly alpha line. All of the properties of the z = 6.5 sample appear to be very similar to those of Ly alpha emitters at lower redshifts., Comment: Accepted for publication in the Astrophysical Journal Letters

We have measured the redshift evolution of the density of Lyman limit systems (LLS) in the intergalactic medium over the redshift range 0 < z < 6. We have used two new quasar samples to (1) improve coverage at z ~ 1, with GALEX grism spectrograph observations of 50 quasars with 0.8 < z_em < 1.3, and (2) extend coverage to z ~ 6, with Keck ESI spectra of 25 quasars with 4.17 < z_em < 5.99. Using these samples together with published data, we find that the number density of LLS per unit redshift, n(z), can be well fit by a simple evolution of the form n(z) = n_3.5 [(1+z)/4.5]^gamma, with n_3.5 = 2.80 +/- 0.33 and gamma = 1.94^(+0.36)_(-0.32) for the entire range 0 < z < 6. We have also reanalyzed the evolution of damped Lyman alpha systems (DLAs) in the redshift range 4 < z < 5 using our high-redshift quasar sample. We find a total of 17 DLAs and sub-DLAs, which we have analyzed in combination with published data. The DLAs with log (HI column density) > 20.3 show the same redshift evolution as the LLS. When combined with previous results, our DLA sample is also consistent with a constant Omega_DLA= 9 x 10^(-4) from z = 2 to z = 5. We have used the LLS number density evolution to compute the evolution in the mean free path of ionizing photons. We find a smooth evolution to z ~ 6, very similar in shape to that of Madau, Haardt & Rees (1999) but about a factor of two higher. Recent theoretical models roughly match to the z < 6 data but diverge from the measured power law at z > 6 in different ways, cautioning against extrapolating the fit to the mean free path outside the measured redshift range., Comment: To be published in ApJ. 30 pages, emulateapj style. Corrects typos in text and tables

Using the extremely high S/N quasar absorption-line sample described in the first paper of the series, we investigate which intergalactic CIV absorption line systems could be directly associated with galactic outflows at z = 2 - 3.5 from an analysis of the velocity widths of the CIV absorption line systems. Only about half the systems with a peak tau(CIV) above 0.4 in the 1548 Angstrom line (roughly a column density of CIV above about 2 x 10^13 cm^-2) have velocity widths large enough to originate in this way, and very few of the weaker systems do. The median velocity full width at a tenth max is found to be 50 km/s for systems with peak tau(CIV) in the range 0.1--0.4 and 160 km/s for systems with a peak tau(CIV) in the range 0.4--3. We show that this critical value of tau(CIV) also separates systems that could be ionized by galaxy-like spectra from those in which the ionization is clearly AGN-dominated. Together the results are consistent with a picture in which almost all the lower column density, and at least half the higher column density, systems lie in the more general IGM whereas about half of the higher column density systems could be produced directly by the outflows and possibly be ionized by their parent galaxies., Comment: 25 pages, 12 figures. Accepted for publication in AJ

We have analyzed the properties of metals in the high redshift intergalactic medium using a novel objective pixel optical depth technique on a sample of extremely high signal-to-noise Keck HIRES and ESI spectra of 26 quasars between redshifts 2.1 and 6.4. The technique relies on using the doublet nature of the common ions CIV and SiIV that are the principal metal tracers in the intergalactic medium outside of the Ly alpha forest. Optical depths are statistically corrected for contamination by other lines, telluric absorption, bad pixels, continuum fitting, etc. and for incompleteness, and we achieve in this way an increased sensitivity of approximately 0.5 dex over previous analyses. Unlike existing POD techniques, we do not compare the ion optical depths with HI optical depths; we therefore avoid problems arising from different velocity widths in the ion and HI. We have shown how the conventional analysis can be reproduced using a percolation method to generate pseudo-clouds from ion optical depths. We find that for the higher resolution HIRES data there is a tight relation, tau ~ N^{0.7}, between the peak optical depth and the column density. From the optical depth vectors themselves we show that there is little evolution in the total amount of CIV from z = 2 to z = 5, though there is a turndown of at least a factor of two in Omega(CIV) above z = 5. We do, however, see substantial evolution in the ratio, SiIV/CIV. Two subsequent papers will investigate what fraction of the absorbers lie in galatic wind outflows (Paper II) and what metallicity is associated with regions of tau(Ly alpha) < 1 (Paper III)., Comment: 38 pages, 23 figures. Accepted for publication in AJ. Spectral data will be available at http://www.ifa.hawaii.edu/~acowie/spectra.html on publication of the paper

We have measured the transmission of the Lyman alpha forest produced by neutral hydrogen scattering in the intergalactic medium between redshifts 2 and 6.3 using high signal to noise, high resolution (R > 5000) observations of 50 quasars spread over the redshift range. We use a uniform set of 15 Angstrom intervals covering Lyman alpha, Lyman beta, and Lyman gamma absorption regions to tabulate the forest transmission as a function of redshift. The transmitted fractions show a relatively smooth evolution over the entire range of redshifts, which can be modelled with a smoothly decreasing ionization rate. Previous claims of an abrupt change at a redshift of approximately 6 appear in part to be a consequence of an incorrect conversion of Lyman beta to Lyman alpha optical depths. The tabulated transmissions can be used to calculate the colors of objects with a specified input spectrum as a function of redshift. We calculate the colors of a flat f_nu galaxy with a large intrinsic continuum break, as an important example., Comment: 32 pages, 6 figures, 6 tables

We report the results of an extensive imaging and spectroscopic survey in the GOODS-North field completed using DEIMOS on the Keck II telescope. Observations of 2018 targets in a magnitude-limited sample of 2911 objects to R=24.4 yield secure redshifts for a sample of 1440 galaxies and AGN plus 96 stars. In addition to redshifts and associated quality assessments, our catalog also includes photometric and astrometric measurements for all targets detected in our R-band imaging survey of the GOODS-North region. We investigate various sources of incompleteness and find the redshift catalog to be 53% complete at its limiting magnitude. The median redshift of z=0.65 is lower than in similar deep surveys because we did not select against low-redshift targets. Comparison with other redshift surveys in the same field, including a complementary Hawaii-led DEIMOS survey, establishes that our velocity uncertainties are as low as 40 km/s for red galaxies and that our redshift confidence assessments are accurate. The distributions of rest-frame magnitudes and colors among the sample agree well with model predictions out to and beyond z=1. We will release all survey data, including extracted 1-D and sky-subtracted 2-D spectra, thus providing a sizable and homogeneous database for the GOODS-North field which will enable studies of large scale structure, spectral indices, internal galaxy kinematics, and the predictive capabilities of photometric redshifts., Comment: 17 pages, 18 figures, submitted to AJ; v2 minor changes; see survey database at http://www2.keck.hawaii.edu/realpublic/science/tksurvey/

We analyze the Ly alpha forest properties of a sample of 15 high redshift quasars lying between z = 4.42 and z = 5.75, using high signal-to-noise spectra obtained with ESI on the Keck II 10 m telescope. The distribution of transmissions in the Ly alpha region in this redshift range is shown to be consistent with that found in lambda cold dark matter simulations with a smoothly evolving ionization rate as a function of redshift. The extrapolation of the ionizing flux to z = 6.05 lies a factor of two higher than a 2 sigma upper limit placed by Cen & McDonald (2001) at this redshift, based on the Becker et al. (2001) spectra of the z = 6.28 quasar SDSS 1030+0524. However, the data are also consistent with models in which there is substantial variation of the ionization parameter about the mean value, and in this case, dark gaps such as those seen by Becker et al. become much more likely. We conclude that further data are needed before we can be sure that we have seen the epoch of reionization. We also summarize the damped Ly alpha systems seen in these quasar lines of sight and measure their metallicities and kinematic properties. We argue that the mean DLA metallicity has dropped substantially by z = 5 compared with its value at z < 4., Comment: 17 pages, including 13 figures. Accepted for publication in Astronomical Journal (May 2002)

It appears that the Lyman alpha forest is becoming thick at a redshift of about 5.5, cutting off the higher redshift intergalactic medium from view in neutral hydrogen. However, the effects of star formation at higher redshift are still readable in the intergalactic metal lines. In this paper I use observations of 32 quasars with emission redshifts in the range 2.31 to 5.86 to study the evolution of the intergalactic metal density from z = 1.5 to z = 5.5. The C IV column density distribution function is consistent with being invariant throughout this redshift range. From direct integration, I determine Omega_CIV to be in the range (2.5 - 7) \times 10^{-8} and Omega_SiIV in the range (0.9 - 3) \times 10^{-8} between z = 1.5 and z = 5. The metallicity at z = 5 exceeds 3.5 \times 10^{-4}, which in turn implies that this fraction of the universal massive star formation took place beyond this redshift. This is sufficient to have ionized the intergalactic medium., Comment: This posting (v3) adds a factor of (1/c) in equation (1), and also the definition of Delta X, both omitted from v2. This is a typographical error and there is no impact on the published values of Omega_ion in the text and figures. v2: Revised to correct an error in equation (1) in the published version. This was a typographical error only and all values of Omega_ion in the text and figures of the published version were correctly computed with H_0 = 65 km/s/Mpc and Omega_m = 1. (5 pages including 4 figures.)

We present an HST STIS spectrum of the HeII Gunn-Peterson effect towards HE2347-4342. Compared to the previous HST GHRS data obtained by Reimers et al. (1997), the STIS spectrum has a much improved resolution. The 2-D detector also allows us to better characterize the sky and dark background. We confirm the presence of two spectral ranges of much reduced opacity, the opacity gaps, and provide improved lower limits on the HeII G-P opacity in the high opacity regions. We use the STIS spectrum together with a Keck--HIRES spectrum covering the corresponding HI Lya forest to calculate a 1-D map of the softness S of the ionization radiation along the line of sight towards HE 2347-4342, where S is the ratio of the HI to HeII photoionization rates. We find that S is generally large but presents important variations, from S ~ 30 in the opacity gaps to a 1 sigma lower limit of 2300 at z~2.86, in a region which shows an extremely low HI opacity over a 6.5 A range. We note that a large S naturally accounts for most of the large SiIV to CIV ratios seen in other quasar absorption line spectra. We present a simple model that reproduces the shape of the opacity gaps in absence of large individual absorption lines. We extend the model described in Heap et al. (2000) to account for the presence of sources close to the line of sight of the background quasar. As an alternative to the delayed reionization model suggested by Reimers et al. (1997), we propose that the large softness observed at z~2.86 is due to the presence of bright soft sources close to the line of sight, i.e. for which the ratio between the number of HI to HeII ionizing photons reaching the IGM is large. We discuss these two models and suggest ways to discriminate between them., Comment: 43 pages, incl. 10 figures. Revised version accepted for publication in ApJ. Tentatively scheduled for the v564 n2 ApJ January 10, 2002 issue. More discussion about the origin of high opacity regions

We have obtained an exceptionally high S/N, high resolution spectrum of the gravitationally lensed quasar Q1422+231 in order to investigate the metal enrichment of the IGM. A column density distribution function (CDDF) is determined for CIV, which is well fit by a power law with index 1.44 +/- 0.05. Once a correction for incompleteness is applied, we find that the CDDF is consistent with a power law down to log N(CIV) = 11.75, a factor of 10 more sensitive than previously determined. In order to determine whether the CIV enrichment extends to even lower column density HI clouds, we find that an analysis of individual pixel optical depths is more robust than the stacking technique. The results obtained from this optical depth method are compared with analyses of simulated spectra enriched with varying CIV enrichment recipes. From these simulations, we conclude that more CIV than is currently directly detected in Q1422+231 is required to reproduce the optical depths determined from the data, consistent with the conclusions drawn from consideration of the power law distribution. (Abridged), Comment: Accepted for publication by AJ. 10 pages of text plus 11 figures and 2 tables

In order to ascertain the extent of metal enrichment in the Ly alpha forest, we have analysed a very high S/N spectrum of the z = 3.625 QSO Q1422+231. We find that in high column density Ly alpha clouds, the power law column density distribution function of C IV continues down to log N(C IV) = 11.7. In addition, by analysing pixel-by-pixel optical depths we show that there are considerably more metals in the Ly alpha forest than are currently directly detectable., Comment: To appear in `Cosmic Evolution and Galaxy Formation: Structure, Interaction and Feedback', ASP conference series, 2000. Ed. J Franco, E. Terlevich, O. Lopez-Cruz and I. Aretxaga

The rate of mass accumulation due to galaxy merging depends on the mass, density, and velocity distribution of galaxies in the near neighborhood of a host galaxy. The fractional luminosity in kinematic pairs combines all of these effects in a single estimator which is relatively insensitive to population evolution. Here we use a k-corrected and evolution compensated volume-limited sample drawing about 300 redshifts from CFGRS and 3000 from CNOC2 to measure the rate and redshift evolution of merging. We identify kinematic pairs with projected separations less than either 50 or 100 \hkpc and rest-frame velocity differences of less than 1000\kms. The fractional luminosity in pairs is modeled as f_L(Delta v,r_p,M_r^{ke})(1+z)^{m_L} where [f_L,m_L] are [0.14+/-0.07,0+/-1.4] and [0.37+/-0.7,0.1+/-0.5] for r_p<= 50 and 100\hkpc, respectively (Omega_M=0.2, Omega_Lambda=0.8). The value of m_L is about 0.6 larger if Lambda=0. To convert these redshift space statistics to a merger rate we use the data to derive a conversion factor to physical space pair density, a merger probability and a mean in-spiral time. The resulting mass accretion rate per galaxy (M_1,M_2>= 0.2 M*) is 0.02+/-0.01(1+z)^{0.1+/-0.5} M*~Gyr^{-1}. Present day high-luminosity galaxies therefore have accreted approximately 0.15M* of their mass over the approximately 7 Gyr to redshift one. (abridged), Comment: accepted for publication in ApJ Letters

A redshift survey has been carried out in the region of the Hubble Deep Field North using the Low Resolution Imaging Spectrograph at the Keck Observatory. The resulting redshift catalog, which contains 671 entries, is a compendium of our own data together with published LRIS/Keck data. It is more than 92% complete for objects, irrespective of morphology, to $R = 24$ mag in the HDF itself and to $R = 23$ mag in the Flanking Fields within a diameter of 8 arcmin centered on the HDF, an unusually high completion for a magnitude limited survey performed with a large telescope. A median redshift $z = 1.0$ is reached at $R \sim 23.8$. Strong peaks in the redshift distribution, which arise when a group or poor cluster of galaxies intersect the area surveyed, can be identified to $z \sim 1.2$ in this dataset. More than 68% of the galaxies are members of these redshift peaks. In a few cases, closely spaced peaks in $z$ can be resolved into separate groups of galaxies that can be distinguished in both velocity and location on the sky. The radial separation of these peaks in the pencil-beam survey is consistent with a characteristic length scale for the their separation of $\approx$70 Mpc in our adopted cosmology ($h = 0.6, \Omega_M = 0.3$, $\Lambda = 0$). Strong galaxy clustering is in evidence at all epochs back to $z \le 1.1$. (abstract abridged), Comment: Accepted to the ApJ. This version contains all the figures and tables. 2 minor typos in table 2b corrected

We report new limits on the Gunn-Peterson effect at a redshift near 5 using spectroscopic observations of the z = 5 Sloan Digital Sky Survey quasar, J033829.31+002156.3, made with the LRIS and HIRES spectrographs on the Keck telescopes. Lower resolution spectrophotometrically calibrated observations made with LRIS over the wavelength region 4500-9600 angstroms were used to obtain a continuum shape and to flux calibrate much higher resolution (R = 36,000) observations made with HIRES. The LRIS data show an Oke D_A index of 0.75. Portions of the HIRES spectrum return to near the extrapolated continuum level. Including both statistical and systematic errors we place an upper limit of tau = 0.1 on the regions of minimum opacity. We argue that, even if this opacity arises in underdense regions of the intergalactic gas, we require a high value of the metagalactic ionizing flux at these redshifts ($J_{\nu} \gg 4 \times 10^{-23}~{\rm ergs cm^{-2} s^{-1} Hz^{-1} sr^{-1}}$ at $z\sim4.72$) to produce a solution which is consistent with even minimum nucleosynthesis estimates of the baryon density. We also report the presence of a Mg II absorption system of extremely high equivalent width ($W_{\lambda,rest}(2796)=1.73$ \AA) at z=2.304., Comment: 15 pages including 2 figures. Also available at http://www.ifa.hawaii.edu/~acowie/gp_5.html To be published in Astrophysical Journal Letters

We have utilized various magnitude-limited samples drawn from an extremely deep and highly complete spectroscopic redshift survey of galaxies observed in seven colors in the Hawaii Survey Fields and the Hubble Deep Field to investigate the evolution of the universal rest-frame ultraviolet luminosity density from z = 1 to the present. The multi-color data (U', B, V, R, I, J, HK') enable the sample selection to be made in the rest-frame ultraviolet for the entire redshift range. Due to the large sample size and depth (U_{AB} = 24.75, B_{AB} = 24.75, I_{AB} = 23.5), we are able to accurately determine the luminosity density to z = 1. We do not confirm the very steep evolution reported by Lilly et al. (1996) but instead find a shallower slope, approximately (1+z)^{1.5} for q0 = 0.5, which would imply that galaxy formation is continuing smoothly to the present time rather than peaking at z = 1. Much of the present formation is taking place in smaller galaxies. Detailed comparisons with other recent determinations of the evolution are presented., Comment: 37 pages including 18 figures. Also available at http://www.ifa.hawaii.edu/~acowie/uvlum.html To be published in the August, 1999 Astronomical Journal (accepted April 22, 1999)

Metal line ratios in a sample of 13 quasar spectra obtained with the HIRES spectrograph on the KeckI telescope have been analyzed to characterize the evolution of the metagalactic ionizing flux near a redshift of 3. The evolution of SiIV/CIV has been determined using three different techniques: using total column densities of absorption line complexes, as in Songaila & Cowie (1996); using the column densities of individual Voigt profile components within complexes; and using direct optical depth ratios. All three methods show that SiIV/CIV changes abruptly at a redshift near 3, requiring a jump in value of about a factor of 3.4, and indicating a significant change in the ionizing spectrum that occurs rapidly between z = 2.9 and z = 3, just above the redshift at which Reimers et al. (1997) detected patchy HeII Lyman alpha absorption. At lower redshifts, the ionization balance is consistent with a pure power law ionizing spectrum but at higher redshifts the spectrum must be very soft, with a large break at the He+ edge. An optical depth ratio technique is used to measure the abundances of ions whose transitions lie within the forest and CIII, SiIII and OVI are detected in this way. The presence of a significant amount of OVI at z > 3 suggests either a considerable volume of HeIII bubbles embedded in the more general region where the ionizing flux is heavily broken, or the addition of collisional ionization to the simple photoionization models., Comment: 51 pages including 21 encapsulated postscript figures. Full version, including complete Figure 5, available at http://www.ifa.hawaii.edu/~acowie/meta_flux.html To be published in the June, 1998 Astronomical Journal (accepted February 18, 1998)

The current status of high redshift D/H measurements is discussed. I first examine whether the observations of HS1937-1009 require a low value of D/H. It is shown that the LRIS measurements of the continuum break and the high resolution Lyman series data can easily be modelled with D/H in the range 5 x 10^(-5) to 10^(-4). I then discuss measurements on weaker Lyman limit systems. A statistical treatment of 10 partial Lyman limit systems favors D/H between 5 x 10^(-5) and 2 x 10^(-4)., Comment: 8 pages LaTeX (aaspp4) with 3 encapsulated postscript figures. Also available at http://www.ifa.hawaii.edu/faculty/acowie/iap.html . To be published in Proceedings of the 13th IAP Colloquium: Structure and Evolution of the IGM from QSO Absorption Lines

Column density distribution functions of CIV with 12.05 < log (N) < 14.35 and SiIV with 11.70 < log (N) < 13.93 have been obtained using 81 CIV absorbers and 35 SiIV absorbers redward of the Ly alpha forest in the lines of sight to seven quasars with 2.518 < z(em) < 3.78. These distribution functions have been directly integrated to yield ion densities at z = 3 to 3.5 of Omega(CIV) = (2.0 \pm 0.5) x 10(-8) and Omega(SiIV) = (7.0 \pm 2.6) x 10(-9) with H0 = 65 km/s/Mpc and q0 = 0.02 (1 sigma errors). A larger sample of 11 quasar lines of sight was used to measure CII/CIV, SiIII/SiIV, and NV/CIV ratios, which suggest that CIV and SiIV are the dominant ionization stages and that corrections to Omega(Carbon) and Omega(Silicon) are no more than a factor of two. Normalizing the alpha-process elements to silicon and the Fe-coproduction elements to carbon gives a density of heavy elements in these forest clouds of Omega(metals) = (3.3 \pm 0.8) x 10(-7) (H0 = 65, q0 = 0.02). The implications for the amount of star formation and for the ionization of the IGM prior to z = 3 are discussed., Comment: 12 pages LaTeX (aaspp4.sty) with 3 encapsulated postscript figures. To be published in ApJ Letters (accepted September 3, 1997)

We have analyzed the properties of low column density Lyman alpha forest clouds (12.8 < log N_HI < 16.0) toward 5 QSOs at different redshifts, using high signal-to-noise data from the HIRES spectrograph on the Keck I 10-m telescope. The results are used to study the evolution of these clouds in the interval 2.1 < z < 3.5. We find: 1) The differential density distribution of forest clouds, f(N_HI), fits an empirical power law with a typical slope of -1.35 to -1.55 for clouds with N_HI < 10^14.3 cm^{-2} and changes extremely slowly with redshift over this N_HI range. 2) The deviation of f(N_HI) from this power-law distribution at higher column densities depends on redshift: for higher redshifts, the departure from a power-law distribution occurs at higher N_HI, which means that higher column density clouds rapidly disappear with decreasing redshift. This may be consistent with an invariant shape for the different column density distributions, but with a shift to a lower column density normalization for systems at different redshifts driven by the overall expansion of the universe. 3) The line widths of the forest clouds increase as redshift decreases. 4) The correlation strength of the forest clouds seems to increase as redshift decreases. We discuss these results in terms of the evolution of the IGM comparing the results to models and analytic descriptions of the evolution of structure in the gas., Comment: 40 pages, LaTeX (aaspp4.sty), includes 4 tables + 18 encapsulated PostScript figures. To appear in The Astronomical Journal, July 1997

A large deep and nearly complete B<24.5 redshift sample is used to measure the change in distribution function of the stellar mass production rate in individual galaxies with redshift. The evolution of the star formation rate distribution with redshift is interpreted in terms of the history of spiral galaxy formation, with the disk component modelled as a single evolving entity, and the characteristic timescales, luminosities, and epochs varying according to galaxy type. The more massive forming galaxies seen at z=1 to 3 are identified as earlier type spirals, whose star formation rates are initially high and then decline rapidly at z<1, while for later type spirals and smaller mass irregulars the mass formation rates at z<1 are lower, and the formation process persists to redshifts much closer to the present epoch. We find that these models can be consistent with the data and fit well into a broad picture of other recent results if qnought=0.02 and many of the disks begin their growth at z<<3, but that they predict too many bright star formers at high z in flat universes., Comment: 12 pages, LaTeX (aaspp4.sty), 3 PostScript figures included. ApJ (Letters) in press

The amount of deuterium relative to hydrogen (D/H) in clouds with close to primordial abundance seen at high redshift in the spectra of distant quasars currently provides the best estimate of the baryonic density of the Universe. The first measurements have yielded discrepant values of D/H both high [around 2 x 10^(-4)] and an order of magnitude lower. The low values of D/H imply a high Omega(Baryon) that is difficult to reconcile with determinations of the primordial abundances of other light elements, notably He^4, and with the known number of light neutrinos. We report an independent measurement of the neutral hydrogen (HI) column density in the cloud toward Q1937-1009 where one of the low D/H values was obtained. Our measurement excludes the reported value and we give a lower limit of D/H > 4 x 10^(-5) in this system, which implies Omega(Baryon) < 0.016 for a Hubble constant of 100 km/sec/Mpc. This new upper limit on Omega(Baryon) relieves the conflict with standard Big Bang nucleosynthesis., Comment: 12 pages LaTeX, including 2 postscript figures. Also available at http://www.ifa.hawaii.edu/faculty/acowie/deut_limits.html . To be published in Nature

We present initial results from a redshift survey carried out with the Low Resolution Imaging Spectrograph on the 10~m W. M. Keck Telescope in the Hubble Deep Field. In the redshift distribution of the 140 extragalactic objects in this sample we find 6 strong peaks, with velocity dispersions of ${\sim}400${\kms}. The areal density of objects within a particular peak, while it may be non-uniform, does not show evidence for strong central concentration. These peaks have characteristics (velocity dispersions, density enhancements, spacing, and spatial extent) similar to those seen in a comparable redshift survey in a different high galactic latitude field (Cohen et al 1996), confirming that the structures are generic. They are probably the high redshift counterparts of huge galaxy structures (``walls'') observed locally., Comment: 14 pages, including 2 figures, to appear in ApJ Letters

We present the results of spectroscopic studies with the LRIS spectrograph on Keck of two of the Hawaii deep survey fields. The 393 objects observed cover an area of 26.2 square arcmin and constitute a nearly complete sample down to K = 20, I = 23, and B = 24.5. The rest-frame K-band luminosity function and its evolution with redshift are described. Comparisons are made with other optically selected (B and I) samples in the literature, and the corresponding rest-frame B-band luminosity function evolution is presented. The B-band counts near B = 24 are shown to be a mixture of normal galaxies at modest redshifts and galaxies undergoing rapid star formation, which have a wide range of masses and which are spread over the redshift interval from z = 0.2 to beyond z = 1.7. The luminosity functions, number counts, and color distributions at optical and IR wavelengths are discussed in terms of a consistent picture of the star-forming history of the galaxy sample. [OII] emission-line diagnostics or rest-frame ultra-violet--infrared color information are used in combination with rest-frame absolute K magnitudes to construct a ``fundamental plane'' in which the evolution of the global star-formation rate with redshift can be shown, and we find that the maximum rest-frame K luminosity of galaxies undergoing rapid star formation has been declining smoothly with decreasing redshift from a value near L* at z > 1. This smooth decrease in the characteristic luminosity of galaxies dominated by star formation can simultaneously account for the high B-band galaxy counts at faint magnitudes and the redshift distribution at z < 1 in both the B- and K-selected samples. Finally, the overall K-band light density evolution is discussed as a tracer of the baryonic mass in stars and compared with the rate of star formation., Comment: 38 pages plus 2 tables, 28 Postscript figures, Figure 1 (finding charts) and Figure 2 (sample spectra) available at http://galileo.ifa.hawaii.edu/~cowie/insight.html . To be published in Astronomical Journal (9/96)

The recent discovery of carbon in close to half of the low neutral hydrogen column density [$N({\rm H~I}) > 3\ten{14}\cm2$] Lyman forest clouds toward $z \sim 3$ quasars has challenged the widely held view of this forest as a chemically pristine population uniformly distributed in the intergalactic medium, but has not eliminated the possibility that a primordial population might be present as well. Using extremely high signal-to-noise observations of a sample of quasars we now show that \ion{C}{4} can be found in 75% of clouds with $N({\rm H~I}) > 3\ten{14}\cm2$ and more than 90% of those with $N({\rm H~I}) > 1.6\ten{15}\cm2$. Clouds with $N({\rm H~I}) > 10^{15}\cm2$ show a narrow range of ionization ratios, spanning less than an order of magnitude in \ion{C}{4}/\ion{H}{1}, \ion{C}{2}/\ion{C}{4}, \ion{Si}{4}/\ion{C}{4} and \ion{N}{5}/\ion{C}{4}, and their line widths require that they be photoionized rather than collisionally ionized. This in turn implies that the systems have a spread of less than an order of magnitude in both volume density and metallicity. Carbon is seen to have a typical abundance of very approximately $10^{-2}$ of solar and Si/C about three times solar, so that the chemical abundances of these clouds are very similar to those of Galactic halo stars. \ion{Si}{4}/\ion{C}{4} decreases rapidly with redshift from high values ($> 0.1$) at $z > 3.1$, a circumstance which we interpret as a change in the ionizing spectrum as the intergalactic medium becomes optically thin to He$^+$\ ionizing photons. Weak clustering is seen in the \ion{C}{4} systems for $\Delta v < 250\kms$, which we argue provides an upper limit to the clustering of \ion{H}{1} clouds. If the clouds are associated with galaxies, this requires a rapid evolution in galaxy clustering between $z = 3$ and $z = 0$., Comment: 31 pages plus 5 tables, 21 Postscript figures, Figures 1 and 2 available at http://www.ifa.hawaii.edu/faculty/acowie/igm_aj.html . To be published in Astronomical Journal

The complex problem of when and how galaxies formed has not until recently been susceptible of direct attack. It has been known for some time that the excessive number of blue galaxies counted at faint magnitudes implies that a considerable fraction of the massive star formation in the universe occurred at z < 3, but, surprisingly, spectroscopic studies of galaxies down to a B magnitude of 24 found little sign of the expected high-z progenitors of current massive galaxies, but rather, in large part, small blue galaxies at modest redshifts z \sim 0.3. This unexpected population has diverted attention from the possibility that early massive star-forming galaxies might also be found in the faint blue excess. From KECK spectroscopic observations deep enough to encompass a large population of z > 1 field galaxies, we can now show directly that in fact these forming galaxies are present in substantial numbers at B \sim 24, and that the era from redshifts 1 to 2 was clearly a major period of galaxy formation. These z > 1 galaxies have very unusual morphologies as seen in deep HST WFPC2 images., Comment: 10 pages LaTeX + 5 PostScript figures in uuencoded gzipped tar file; aasms4.sty, flushrt.sty, overcite.sty (the two aastex4.0 and overcite.sty macros are available from xxx.lanl.gov) Also available (along with style files) via anonymous ftp to ftp://hubble.ifa.hawaii.edu/pub/preprints . E-print version of paper adds citation cross-references to other archived e-prints, where available. To appear in Nature October 19, 1995

We report new upper limits on any possible long-term time variation of the ratio of the electron to proton masses, $(m_e/m_p)$, the fine-structure constant, ($\alpha$), and the quantity $\alpha^2 g_p(m_e/m_p)$, where $g_p$ is the proton gyromagnetic ratio. These limits are based on extremely high precision observations of H$_2$, Si$^{3+}$, C$^0$ and H$^0$ in high-redshift quasar absorption lines. They amount to 95\% confidence ranges of $(-7.6\to 9.7) \times 10^{-14}$ yr$^{-1}$ for ($m_e/m_p$), $(-4.6\to 4.2)\times 10^{-14}$ yr$^{-1}$ for $\alpha$\ and $(-2.2\to 4.2) \times 10^{-15}$ yr$^{-1}$ for $\alpha^2 g_p(m_e/m_p)$, where the elapsed time has been computed for a cosmology with \hnought = 75 km s$^{-1}$ Mpc$^{-1}$ and \qnought = 0.5., Comment: 10 pages (w/ 2 tables) of text as 1 LaTeX file (aastex style macros: aaspp.sty, flushrt.sty) plus 1 uuencoded compressed tar file of 2 PostScript figures. To appear in the November 10, 1995 Astrophysical Journal

We present very deep $HST$ WFPC2 images in the F814W filter of two Hawaii Survey fields, SSA13 and SSA22. Using these data with previous ground-based imaging and spectroscopy, we compare the colors, star-forming properties and morphologies of the faintest galaxies with a reference sample of bright nearby galaxies and analyze the changes in field galaxy morphology with magnitude. Our principal result is the identification of a new morphological class of ``chain'' galaxies at the faintest magnitudes. Based on limited spectroscopy, we tentatively conclude that these are linearly organized giant star-forming regions at $z = 0.5-3$ and, if this is correct, that these are large galaxies in the process of formation., Comment: 18 pages + 1 table of text as 1 LaTeX file (uses aastex style macros: aaspp.sty, flushrt.sty) plus 1 uuencoded compressed tar file of 12 PostScript figures (Figs. 3-9, 16-17, and 21-23). The remaining gray-scale plots are available by anonymous ftp at ftp://hubble.ifa.hawaii.edu/pub/preprints/plates To appear in the October 1995 Astronomical Journal

We describe the properties of the Ly$\alpha$ forest in the column density range $\nhi \geq 2\times10^{12}$ cm$^{-2}$ based on 1056 lines in the wavelength range 4300--5100\AA measured in extremely high S/N, $R=36,000$ spectra of four quasars. The column density distribution is well described by a -1.5 power law to $2\times10^{12}$ cm$^{-2}$, below which limit confusion becomes too severe to measure a spectrum of individual clouds. The distribution of $b$ values shows a well-defined lower envelope with a cutoff at $b=20\kms$ corresponding to a cloud temperature of 24,000 K. There is only a very small fraction (less than 1%) of narrow line clouds which cannot be identified with metal-lines. From modeling the Ly$\alpha$ absorption lines as complexes of clouds each with thermal broadening corresponding to $b_c$ we find the $b$ distribution can be understood if there is a mean of 3.25 clouds per absorption line with a spread in velocity centroids characterized by a dispersion of $10.75\kms$., Comment: 17 pages + 3 tables of text as 1 LaTeX file (uses aastex version 4 style macros: aaspp4.sty, flushrt) plus 1 uuencoded compressed tar file of 7 PostScript figures. Appendix tables and figures [complete spectra and line lists (~1.6 Mb uncompressed)] are available by anonymous ftp at ftp://hubble.ifa.hawaii.edu/pub/preprints. To appear in the October 1995 Astronomical Journal

Recent progress in the characterization of the interstellar medium (ISM) by means of optical and UV spectral data is summarized. The gas is studied by focusing on background stars whose spectra can be accurately modeled to provide the light source for the absorption-line scans. The capabilities of earth- and space-based instruments which have been and are used for the surveys are delineated. The distributions of diffuse gas densities and characteristics of the cold, warm and hot gas in the Galaxy are described in terms of the elemental abundances, kinetics and distributions of the gas. Particular note is taken of gas in the solar neighborhood and around SNR, and of absorption-line data of cosmological significance.