arXiv subjects

Lennox L. Cowie

Publications and source records attributed to Lennox L. Cowie.

At least 37 records · Page 2Linked to original sources

arXiv2015

Luminosity Dependence and Redshift Evolution of Strong Emission-line Diagnostics in Star-Forming Galaxies

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.

astro-ph.GA↗

arXiv2014

Local Large-Scale Structure and the Assumption of Homogeneity

Our recent estimates of galaxy counts and the luminosity density in the near-infrared (Keenan et al. 2010, 2012) indicated that the local universe may be under-dense on radial scales of several hundred megaparsecs (Mpc). Such a large-scale local under-density could introduce significant biases in the measurement and interpretation of cosmological observables, such as the inferred effects of dark energy on the rate of expansion. In Keenan et al. (2013), we measured the K-band luminosity density as a function of distance from us to test for such a local under-density. We made this measurement over the redshift range 0.01 < z < 0.2 (radial distances D ~ 50-800 Mpc). We found that the shape of the K-band luminosity function is relatively constant as a function of distance and environment. We derive a local (z < 0.07, D < 300 Mpc) K-band luminosity density that agrees well with previously published studies. At z > 0.07, we measure an increasing luminosity density that by z~ 0.1 rises to a value of ~1.5 times higher than that measured locally. This implies that the stellar mass density follows a similar trend. Assuming that the underlying dark matter distribution is traced by this luminous matter, this suggests that the local mass density may be lower than the global mass density of the universe at an amplitude and on a scale that is sufficient to introduce significant biases into the measurement of basic cosmological observables. At least one study has shown that an under-density of roughly this amplitude and scale could resolve the apparent tension between direct local measurements of the Hubble constant and those inferred by Planck team. Other theoretical studies have concluded that such an under-density could account for what looks like an accelerating expansion, even when no dark energy is present.

astro-ph.CO↗

arXiv2014

SMA observations on faint Submillimeter Galaxies with S$_{850}$ < 2 mJy: Ultra Dusty Low-Luminosity Galaxies at High Redshift

We obtained SMA observations of eight faint (intrinsic 850 $μ$m fluxes $<$ 2 mJy) submillimeter galaxies (SMGs) discovered in SCUBA images of the massive lensing cluster fields A370, A2390, and A1689 and detected five. In total, we obtain 5 SMA detections, all of which have de-lensed fluxes $<$1 mJy with estimated total infrared luminosities 10$^{10}-10^{12}$ $L_\odot$, comparable to luminous infrared galaxies (LIRGs) and normal star-forming galaxies. Based on the latest number counts, these galaxies contribute $\sim$70% of the 850 $μ$m extragalactic background light and represent the dominant star-forming galaxy population in the dusty universe. However, only 40$^{+30}_{-16}$% of our faint SMGs would be detected in deep optical or near-infrared surveys, which suggests many of these sources are at high redshifts ($z \gtrsim 3$) or extremely dusty, and they are not included in current star formation history estimates.

astro-ph.CO↗

arXiv2014

z~1 Lya Emitters I. The Luminosity Function

We construct a flux-limited sample of 135 candidate z~1 Lya emitters (LAEs) from Galaxy Evolution Explorer (GALEX) grism data using a new data cube search method. These LAEs have luminosities comparable to those at high redshifts and lie within a 7 Gyr gap present in existing LAE samples. We use archival and newly obtained optical spectra to verify the UV redshifts of these LAEs. We use the combination of the GALEX UV spectra, optical spectra, and X-ray imaging data to estimate the active galactic nucleus (AGN) fraction and its dependence on Lya luminosity. We remove the AGNs and compute the luminosity function (LF) from 60 z~1 LAE galaxies. We find that the best fit LF implies a luminosity density increase by a factor of ~1.5 from z~0.3 to z~1 and ~20 from z~1 to z~2. We find a z~1 volumetric Lya escape fraction of 0.7+/-0.4%.

astro-ph.CO↗

arXiv2013

Resolving the Cosmic Far-Infrared Background at 450 and 850 Microns with SCUBA-2

We use the SCUBA-2 submillimeter camera mounted on the JCMT to obtain extremely deep number counts at 450 and 850um. We combine data on two cluster lensing fields, A1689 and A370, and three blank fields, CDF-N, CDF-S, and COSMOS, to measure the counts over a wide flux range at each wavelength. We use statistical fits to broken power law representations to determine the number counts. This allows us to probe to the deepest possible level in the data. At both wavelengths our results agree well with the literature in the flux range over which they have been measured, with the exception of the 850um counts in CDF-S, where we do not observe the counts deficit found by previous single-dish observations. At 450um, we detect significant counts down to ~1mJy, an unprecedented depth at this wavelength. By integrating the number counts above this flux limit, we measure 113.9^{+49.7}_{-28.4} Jydeg^{-2} of the 450um extragalactic background light (EBL). The majority of this contribution is from sources with S_450um between 1-10mJy, and these sources are likely to be the ones that are analogous to the local luminous infrared galaxies (LIRGs). At 850um, we measure 37.3^{+21.1}_{-12.9} Jydeg^{-2} of the EBL. Because of the large systematic uncertainties on the COBE measurements, the percentage of the EBL we resolve could range from 48%-153% (44%-178%) at 450 (850)um. Based on high-resolution SMA observations of around half of the 4sigma 850um sample in CDF-N, we find that 12.5^{+12.1}_{-6.8}% of the sources are blends of multiple fainter sources. This is a low multiple fraction, and we find no significant difference between our original SCUBA-2 850um counts and the multiplicity corrected counts.

astro-ph.CO↗

arXiv2013

Evidence for a ~300 Mpc Scale Under-density in the Local Galaxy Distribution

Galaxy counts and recent measurements of the luminosity density in the near-infrared (NIR) have indicated the possibility that the local universe may be under-dense on scales of several hundred megaparsecs. The presence of a large-scale under-density in the local universe could introduce significant biases into the interpretation of cosmological observables, and, in particular, into the inferred effects of dark energy on the expansion rate. Here we measure the K-band luminosity density as a function of redshift to test for such a local under-density. In this effort, we combine photometry from UKIDSS and 2MASS with redshifts from the SDSS, 2DFGRS, 6DFGRS, 2MR, and GAMA surveys. We find that the overall shape of the z=0 rest-frame K-band luminosity function (M* = -22.15 +/- 0.04 and alpha = -1.02 +/- 0.03) appears to be relatively constant as a function of environment and redshift out to z ~0.2. We find a local (z < 0.07) luminosity density that is in good agreement with previous studies. Beyond z ~ 0.07 we detect a rising luminosity density that reaches a value ~1.5 times higher than that measured locally at z>0.1. This suggests that the stellar mass density as a function of redshift follows a similar trend. Assuming that luminous matter traces the underlying dark matter distribution, this implies that the local mass density of the universe may be lower than the global value on a scale and amplitude sufficient to introduce significant biases into the determination of basic cosmological observables, such as the expansion rate. An under-density of roughly this scale and amplitude would be sufficient to resolve the apparent tension between direct measurements of the Hubble constant and those inferred by Planck.

astro-ph.CO↗

arXiv2013

A SCUBA-2 850 micron survey of protoplanetary discs in the sigma Orionis cluster

We present the results from a large 850 micron survey of the sigma Orionis cluster using the SCUBA-2 camera on the James Clerk Maxwell Telescope. The 0.5-degree diameter circular region we surveyed contains 297 young stellar objects with an age estimated at about 3Myr. We detect 9 of these objects, 8 of which have infrared excesses from an inner disc. We also serendipitously detect 3 non-stellar sources at > 5sigma that are likely background submillimetre galaxies. The 9 detected stars have inferred disc masses ranging from 5 to about 17MJup, assuming similar dust properties as Taurus discs and an ISM gas-to-dust ratio of 100. There is a net positive signal toward the positions of the individually undetected infrared excess sources indicating a mean disc mass of 0.5 MJup . Stacking the emission toward those stars without infrared excesses constrains their mean disc mass to less than 0.3MJup, or an equivalent Earth mass in dust. The submillimetre luminosity distribution is significantly different from that in the younger Taurus region, indicating disc mass evolution as star forming regions age and the infrared excess fraction decreases. Submillimeter Array observations reveal CO emission toward 4 sources demonstrating that some, but probably not much, molecular gas remains in these relatively evolved discs. These observations provide new constraints on the dust and gas mass of protoplanetary discs during the giant planet building phase and provide a reference level for future studies of disc evolution.

astro-ph.GA↗

arXiv2012

Faint Submillimter Galaxy Counts at 450 micron

We present the results of SCUBA2 observations at 450 micron and 850 micron of the field lensed by the massive cluster A370. With a total survey area > 100 arcmin2 and 1 sigma sensitivities of 3.92 and 0.82 mJy/beam at 450 and 850 micron respectively, we find a secure sample of 20 sources at 450 micron and 26 sources at 850 micron with a signal-to-noise ratio > 4. Using the latest lensing model of A370 and Monte Carlo simulations, we derive the number counts at both wavelengths. The 450 micron number counts probe a factor of four deeper than the counts recently obtained from the Herschel Space Telescope at similar wavelengths, and we estimate that ~47-61% of the 450 micron extragalactic background light (EBL) resolved into individual sources with 450 micron fluxes greater than 4.5 mJy. The faint 450 micron sources in the 4 sigma sample have positional accuracies of 3 arcseconds, while brighter sources (signal-to-noise > 6 sigma) are good to 1.4 arcseconds. Using the deep radio map (1 sigma ~ 6 uJy) we find that the percentage of submillimeter sources having secure radio counterparts is 85% for 450 micron sources with intrinsic fluxes > 6 mJy and 67% for 850 micron sources with intrinsic fluxes > 4 mJy. We also find that 67% of the > 4 sigma 450 micron sources are detected at 850 micron, while the recovery rate at 450 micron of > 4 sigma 850 micron sources is 54%. Combined with the source redshifts estimated using millimetric flux ratios, the recovered rate is consistent with the scenario where both 450 micron and 20 cm emission preferentially select lower redshift dusty sources, while 850 micron emission traces a higher fraction of dusty sources at higher redshifts. [Abridge]

astro-ph.CO↗

arXiv2011

Faint Submillimeter Galaxies behind the Massive Lensing Cluster A2390

Current studies on Submillimeter Galaxies (SMGs) mostly focus on bright sources with 850 micron flux greater than 2 mJy, and the results have shown that they are likely high redshift mergers with z > 2 and could be a dominant population on star formation in the early Universe. However, bright SMGs only contributes 20-30% of the 850 micron extragalactic background light (EBL), meaning the bulk of the cosmic star formation still hidden by dust and our current understanding is biased. We have started a program to study an unbiased sample of highly-amplified and intrinsically faint SCUBA detected SMGs in the field of massive lensing clusters. Here we report the newly obtained SMA observations at 850 micron on one of our sample source, A2390-5, behind the massive lensing cluster A2390. We successfully detect the source with a flux of 3.95 mJy. Surprisingly, it does not have any counterpart in any other wavelengths even though there are tentative candidates, which implies a very dusty and high-z nature. With less than 1" positional accuracy and the adoption of z = 5, we obtain the amplification factor of 12 using current lensing model, which makes A2390-5 a faint SMG with a de-lensed flux of 0.33 mJy. Together with our previous detection on another faint SMG, both of them have no counterpart in other wavelengths and their properties are very different than previously thought from the single-dish data. We emphasize the importance of direct submillimeter high-resolution studies on faint SMGs, which could be the dominant population of the high-z star formation.

astro-ph.CO↗

arXiv2011

A Ks and IRAC Selection of High-Redshift Extremely Red Objects

In order to find the most extreme dust-hidden high-redshift galaxies, we select 196 extremely red objects in the Ks and IRAC bands (KIEROs, [Ks-4.5um](AB)>1.6) in the 0.06 deg^2 GOODS-N region. This selection avoids the Balmer breaks of galactic spectra at z<4 and picks up red galaxies with strong dust extinction. The photometric redshifts of KIEROs are between 1.5 and 5, with ~70% at z~2-4. KIEROs are very massive, with M*~10^10-10^12 Msun. They are optically faint and usually cannot be picked out by the Lyman break selection. On the other hand, the KIERO selection includes approximately half of the known millimeter and submillimeter galaxies in the GOODS-N. Stacking analyses in the radio, millimeter, and submillimeter all show that KIEROs are much more luminous than average 4.5um selected galaxies. Interestingly, the stacked fluxes for ACS-undetected KIEROs in these wavebands are 2.5-5 times larger than those for ACS-detected KIEROs. With the stacked radio fluxes and the local radio-FIR correlation, we derive mean infrared luminosities of 2-7x10^12 Lsun and mean star formation rates of 300-1200 Msun/yr for KIEROs with redshifts. We do not find evidence of a significant subpopulation of passive KIEROs. The large stellar masses and star formation rates imply that KIEROs are $z>2$ massive galaxies in rapid formation. Our results show that a large sample of dusty ultraluminous sources can be selected in this way and that a large fraction of high-redshift star formation is hidden by dust.

astro-ph.CO↗

arXiv2011

The UV-continuum properties of Ly alpha selected galaxies at z=6.5

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.

astro-ph.CO↗

arXiv2011

Ly alpha emitting galaxies as early stages in galaxy formation

We present optical spectroscopy of two samples of GALEX grism selected Ly alpha emitters (LAEs): one at z=0.195-0.44 and the other at z=0.65-1.25. We have also observed a comparison sample of galaxies in the same redshift intervals with the same UV magnitude distributions but with no detected Ly alpha. We use the optical spectroscopy to eliminate active galactic nuclei (AGNs) and to obtain the optical emission-line properties of the samples. We compare the luminosities of the LAEs in the two redshift intervals and show that there is dramatic evolution in the maximum Ly alpha luminosity over z=0-1. Focusing on the z=0.195-0.44 samples alone, we show that there are tightly defined relations between all of the galaxy parameters and the rest-frame equivalent width (EW) of H alpha. The higher EW(H alpha) sources all have lower metallicities, bluer colors, smaller sizes, and less extinction, consistent with their being in the early stages of the galaxy formation process. We find that 75 +- 12% of the LAEs have EW(H alpha)>100 Angstrom, and, conversely, that 31 +/- 3% of galaxies with EW(H alpha)>100 Angstrom are LAEs. We correct the broadband magnitudes for the emission-line contributions and use spectral synthesis fits to estimate the ages of the galaxies. We find a median age of 1.1x10^{8} yr for the LAE sample and 1.4x10^{9} yr for the UV-continuum sample without detected Ly alpha. The median metallicity of the LAE sample is 12+log(O/H)=8.24, or about 0.4 dex lower than the UV-continuum sample.

astro-ph.CO↗

arXiv2011

Submillimeter sources behind the massive lensing clusters A370 and A2390

We report 850 micron Submillimeter Array (SMA) observations of four gravitationally lensed submillimeter galaxies (SMGs), A370-2, A2390-1, A2390-3 and A2390-4, which were originally discovered with the Submillimeter Common-User Bolometer Array (SCUBA). Our SMA detection of A370-2 with a submillimeter flux of 7.95 +/- 0.60 mJy unambiguously identifies the counterparts to this source at optical and radio wavelengths. A2390-1 is an ultraluminous infrared galaxies with a submillimeter flux of 5.55 +/- 0.92 mJy and a redshift of 1.8 +/- 0.2 computed from submillimeter/radio flux ratio analysis. We resolve A2390-3 into two components, A2390-3a and A2390-3b, with fluxes of 3.15 +/- 0.63 mJy and 1.92 +/- 0.60 mJy, respectively. The structure of the system could be consistent with morphological distortion by gravitational lensing. The lack of counterparts in the optical and infrared indicates a heavily dust-enshrouded nature, and a non-detection in the radio implies that these two sources probably lie at z > 4.7, which would make them among the most distant SMGs known to date. Our non-detection of A2390-4 suggests either that there are multiple fainter submillimeter sources within the SCUBA beam or that the SCUBA detection may have been false. Our precise positions allow us to determine accurate amplifications and fluxes for all our detected sources. Our new results give a shallower power-law fit (-1.10) to the faint-end 850 micron cumulative number counts than previous work. We emphasize the need for high-resolution observations of single-dish detected SMGs in order to measure accurately the faint end of the 850 micron counts.

astro-ph.CO↗

arXiv2010

SMAObservations of GOODS 850-11 and GOODS 850-13 - First Examples of Multiple Submillimeter Sources Resolved by an Interferometer

We carried out extremely sensitive Submillimeter Array (SMA) 340 GHz continuum imaging on two submillimeter galaxies (SMGs): GOODS 850-11 and GOODS 850-13. The observations reach sub-mJy rms sensitivities and, interestingly, resolve both sources into multiple, physically unrelated SMGs. GOODS 850-11 is resolved into two sources at different redshifts. GOODS 850-13 is resolved into three sources, two with different spectroscopic redshifts and one only with a photometric redshift. All the SMA sources have fluxes in the 3-5 mJy range and all are detected at 1.4 GHz. Three of them are detected by Chandra, and one is a previously unknown X-ray SMG. This is the first time that single-dish SMGs are resolved into multiple unrelated sources and also the first time that the SMA has discovered new SMGs. Our results show that identifications of SMGs at any wavelengths other than the submillimeter itself can be misleading, since such identifications usually only pick up one of the real counterparts. Using simulations that mimic our SCUBA and SMA observations, we find that the number of triple systems detected in our SMA survey is much higher than that expected from the current best-determined number counts. We tentatively attribute this to clustering. We also predict that ALMA will find ~1/3 of >5 mJy 850 um SCUBA sources to be multiple systems. Based on our SMA observations and simulations, we suggest that large samples of existing SMGs should be imaged with sensitive interferometric observations, even if the SMGs were previously thought to be securely identified.

astro-ph.CO↗

arXiv2010

The Evolution of Lyman Limit Absorption Systems to Redshift Six

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.

astro-ph.CO↗

arXiv2010

The Resolved Near-Infrared Extragalactic Background

We present a current best estimate of the integrated near-infrared (NIR) extragalactic background light (EBL) attributable to resolved galaxies in J, H, and Ks. Our results in units of nW m-2 sr-1 are 11.7+5.6 -2.6 in J, 11.5+4.5 -1.5 in H and 10.0+2.8 -0.8 in Ks. We derive these new limits by combining our deep wide-field NIR photometry from five widely separated fields with other studies from the literature to create a galaxy counts sample that is highly complete and has good counting statistics out to JHKs ~ 27-28. As part of this effort we present new ultradeep Ks-band galaxy counts from 22 hours of observations with the Multi Object Infrared Camera and Spectrograph (MOIRCS) instrument on the Subaru Telescope. We use this MOIRCS Ks-band mosaic to estimate the total missing flux from sources beyond our detection limits. Our new limits to the NIR EBL are in basic agreement with, but 10 - 20% higher than previous estimates, bringing them into better agreement with estimates of the total NIR EBL (resolved + unresolved sources) obtained from TeV gamma-ray opacity measurements and recent direct measurements of the total NIR EBL. We examine field to field variations in our photometry to show that the integrated light from galaxies is isotropic to within uncertainties, consistent with the expected large-scale isotropy of the EBL. Our data also allow for a robust estimate of the NIR light from Galactic stars, which we find to be 14.7 +/- 2.4 in J, 10.1 +/- 1.9 in H and 7.6 +/- 1.8 in Ks in units of nW m-2 sr-1.

astro-ph.CO↗

arXiv2010

Ultradeep Ks Imaging in the GOODS-N

We present an ultradeep Ks-band image that covers 0.5*0.5 deg^2 centered on the Great Observatories Origins Deep Survey-North (GOODS-N). The image reaches a 5 σdepth of Ks(AB) = 24.45 in the GOODS-N region, which is as deep as the GOODS-N Spitzer Infrared Array Camera (IRAC) 3.6 μm image. We present a new method of constructing IRAC catalogs that uses the higher spatial resolution Ks image and catalog as priors and iteratively subtracts fluxes from the IRAC images to estimate the IRAC fluxes. Our iterative method is different from the χ^2 approach adopted by other groups. We verified our results using data taken in two different epochs of observations, as well as by comparing our colors with the colors of stars and with the colors derived from model spectral energy distributions (SEDs) of galaxies at various redshifts. We make available to the community our WIRCam Ks-band image and catalog (94951 objects in 0.25 deg^2), the Interactive Data Language (IDL) pipeline used for reducing the WIRCam images, and our IRAC 3.6 to 8.0 μm catalog (16950 objects in 0.06 deg^2 at 3.6 μm). With this improved Ks and IRAC catalog and a large spectroscopic sample from our previous work, we study the color-magnitude and color-color diagrams of galaxies. We compare the effectiveness of using Ks and IRAC colors to select active galactic nuclei (AGNs) and galaxies at various redshifts. We also study a color selection of z = 0.65--1.2 galaxies using the Ks, 3.6 μm, and 4.5 μm bands.

astro-ph.CO↗

arXiv2009

Deep Spectroscopy of Ultra-Strong Emission Line Galaxies

Ultra strong emission-line galaxies (USELs) with extremely high equivalent widths (EW(H beta) > 30A) can be used to pick out galaxies of extremely low metallicity in the z=0-1 redshift range. Large numbers of these objects are easily detected in deep narrow band searches and, since most have detectable [OIII] 4363, their metallicities determined using the direct method. These large samples hold the possibility for determining if there is a metallicity floor for the galaxy population. Here we describe results of an extensive spectroscopic follow-up of the Kakazu et al. (2007) catalog of 542 USELs using the DEIMOS spectrograph on Keck, with high S/N spectra of 348 galaxies. The two lowest metallicity galaxies in our sample have 12+log(O/H)=6.97+/-0.17 and 7.25+/-0.03 -- values comparable to the lowest metallicity galaxies found to date. We determine an empirical metallicity-R23 parameter relation for our sample, and compare this to the relationship for low redshift galaxies. The determined metallicity-luminosity relation is compared with those of magnitude selected samples in the same redshift range. The emission line selected galaxies show a metal-luminosity relation where the metallicity decreases with luminosity and they appear to define the lower bound of the galaxy metallicity distribution at a given continuum luminosity. We also compute the H alpha luminosity function of the USELs as a function of redshift and use this to compute an upper bound on the Ly alpha emitter luminosity function over the z=0-1 redshift range.

astro-ph.CO↗