Searcharxiv⌕ Search

arXiv subjects

George D. Becker

Publications and source records attributed to George D. Becker.

97 records · Page 6Linked to original sources

High Redshift Metals I.: The Decline of C IV at z > 5.3

We present the results from our search for C IV in the intergalactic medium at redshifts z=5.3-6.0. We have observed four z~6 QSOs with Keck/NIRSPEC in echelle mode. The data are the most sensitive yet taken to search for C IV at these redshifts, being 50% complete at column densities log(N_{CIV})=13.4. We find no C IV systems in any of the four sightlines. Taking into account our completeness, this translates into a decline in the number density of C IV absorbers in the range 13.0 < log(N_{CIV}) < 15.0 of at least a factor 4.4 (95% confidence) from z~2-4.5, where the number density is relatively constant. We use our lack of detections to set limits on the slope and normalization of the column density distribution at z=5.3-6.0. The rapid evolution of C IV at these redshifts suggests that the decrease in the number density may largely be due to ionization effects, in which case many of the metals in the z~4.5 IGM could already be in place at z~5.3, but in a lower ionization state. The lack of weak systems in our data, combined with the presence of strong C IV absorbers along at least one other sightline, further suggests that there may be large-scale variations in the enrichment and/or ionization state of the z~6 IGM, or that C IV absorbers at these redshifts are associated with rare, UV-bright star-forming galaxies.

astro-ph↗

The Evolution of Optical Depth in the Ly-alpha Forest: Evidence Against Reionization at z~6

We examine the evolution of the IGM Ly-alpha optical depth distribution using the transmitted flux probability distribution function (PDF) in a sample of 63 QSOs spanning absorption redshifts 1.7 < z < 5.8. The data are compared to two theoretical optical depth distributions: a model distribution based on the density distribution of Miralda-Escude et al. (2000) (MHR00), and a lognormal distribution. We assume a uniform UV background and an isothermal IGM for the MHR00 model, as has been done in previous works. Under these assumptions, the MHR00 model produces poor fits to the observed flux PDFs at redshifts where the optical depth distribution is well sampled, unless large continuum corrections are applied. However, the lognormal optical depth distribution fits the data at all redshifts with only minor continuum adjustments. We use a simple parametrization for the evolution of the lognormal parameters to calculate the expected mean transmitted flux at z > 5.4. The lognormal optical depth distribution predicts the observed Ly-alpha and Ly-beta effective optical depths at z > 5.7 while simultaneously fitting the mean transmitted flux down to z = 1.6. If the evolution of the lognormal distribution at z < 5 reflects a slowly-evolving density field, temperature, and UV background, then no sudden change in the IGM at z ~ 6 due to late reionization appears necessary. We have used the lognormal optical depth distribution without any assumption about the underlying density field. If the MHR00 density distribution is correct, then a non-uniform UV background and/or IGM temperature may be required to produce the correct flux PDF. We find that an inverse temperature-density relation greatly improves the PDF fits, but with a large scatter in the equation of state index. [Abridged]

astro-ph↗

Discovery of excess O I absorption towards the z = 6.42 QSO SDSS J1148+5251

We present a search for O I in the spectra of nine 4.9 < z_qso < 6.4 QSOs taken with Keck/HIRES. We detect six systems with N(O I) > 10^13.7 cm^{-2} in the redshift intervals where O I 1302 falls redward of the Ly-alpha forest. Four of these lie towards SDSS J1148+5251 (z_qso = 6.42). This imbalance is unlikely to arise from variations in sensitivity among our data or from a statistical fluctuation. The excess O I occurs over a redshift interval that also contains transmission in Ly-alpha and Ly-beta. Therefore, if these O I systems represent pockets of neutral gas, then they must occur within or near regions of the IGM that are highly ionized. In contrast, no O I is detected towards SDSS J1030+0524 (z_qso = 6.30), whose spectrum shows complete absorption in Ly-alpha and Ly-beta over Δz ~ 0.2. Assuming no ionization corrections, we measure mean abundance ratios <[O/Si]> = -0.04 +/- 0.06, <[C/O]> = -0.31 +/- 0.09, and <[C/Si]> = -0.34 +/- 0.07 (2 sigma), which are consistent with enrichment dominated by Type II supernovae. The O/Si ratio limits the fraction of silicon in these systems contributed by metal-free very massive stars to < 30%, a result which is insensitive to ionization corrections. The ionic comoving mass densities along the z_qso > 6.2 sightlines, including only the detected systems, are Ω(O I) = (7.0 +/- 0.6) * 10^{-8}, Ω(Si II) = (9.6 +/- 0.9) * 10^{-9}, and Ω(C II) = (1.5 +/- 0.2) * 10^{-8}.

astro-ph↗

Expansion and Collapse in the Cosmic Web

We study the kinematics of the gaseous cosmic web at high redshift with Lyman alpha forest absorption in multiple QSO sightlines. Using a simple analytic model and a cosmological hydrodynamic simulation we constrain the underlying three-dimensional distribution of velocities from the observed line-of-sight distribution of velocity shear across the plane of the sky. The distribution is found to be in good agreement with the intergalactic medium (IGM) undergoing large scale motions dominated by the Hubble flow. Modeling the Lyman alpha clouds analytically and with a hydrodynamics simulation, the average expansion velocity of the gaseous structures causing the Lyman alpha forest in the lower redshift (z = 2) sample appears about 20 percent lower than the local Hubble expansion velocity. We interpret this as tentative evidence for some clouds undergoing gravitational collapse. However, the distribution of velocities is highly skewed, and the majority of clouds at redshifts from 2 to 3.8 expand typically about 5 - 20 percent faster than the Hubble flow. This behavior is explained if most absorbers in the column density range typically detectable are expanding filaments that stretch and drain into more massive nodes. We find no evidence for the observed distribution of velocity shear being significantly influenced by processes other than Hubble expansion and gravitational instability, like galactic winds. To avoid overly disturbing the IGM, winds may be old and/or limp by the time we observe them in the Lyman alpha forest, or they may occupy only an insignificant volume fraction of the IGM. (abridged)

astro-ph↗

Observations of Chemically Enriched QSO Absorbers near z ~ 2.3 Galaxies: Galaxy-Formation Feedback Signatures in the IGM

We present a study of galaxies and intergalactic gas toward the z=2.73 quasar HS1700+6416, to explore the effects of galaxy formation feedback on the IGM. Our observations and ionization simulations indicate that the volume within 100-200 h_71^{-1} physical kpc of high-redshift galaxies contains very small, dense, and metal-rich absorption-line regions. These systems often contain shock-heated gas seen in OVI, and may exhibit [Si/C] abundance enhancements suggestive of Type II supernova enrichment. We argue that the absorbers resemble thin sheets or bubbles, whose physical properties can be explained with a simple model of radiatively efficient shocks propegating through the IGM. Their high metallicities suggest that these shocks are being expelled from--rather than falling into--star forming galaxies. There is a dropoff in the IGM gas density at galaxy impact parameters beyond ~300 physical kpc that may trace boundaries of gas structures where the galaxies reside. The local heavy-element enhancement covers 100-200 kpc; beyond this the observed abundances blend into the general IGM. Supernova-driven winds or dynamical stripping of interstellar gas appears to affect the IGM near massive galaxies, even at R>~100 kpc. However, these feedback systems represent only a few percent of the Lya forest mass at z~2.5. Their mass could be larger if the more numerous metal-poor CIV systems at >~200 kpc are tepid remnants of very powerful winds. Based on present observations it is not clear that this scenario is to be favored over one involving pre-enrichment by smaller galaxies at z>~6.

astro-ph↗

Large-Scale Correlations in the Lyman-alpha Forest at z = 3-4

We present a study of the spatial coherence of the intergalactic medium toward two pairs of high-redshift quasars with moderate angular separations observed with Keck/ESI, Q1422+2309A/Q1424+2255 (z_em = 3.63, theta = 39") and Q1439-0034A/B (z_em = 4.25, theta = 33"). The crosscorrelation of transmitted flux in the Lyman-alpha forest shows a 5-7 sigma peak at zero velocity lag for both pairs. This strongly suggests that at least some of the absorbing structures span the 230-300/h_70 proper kpc transverse separation between sightlines. We also statistically examine the similarity between paired spectra as a function of transmitted flux, a measure which may be useful for comparison with numerical simulations. In investigating the dependence of the correlation functions on spectral characteristics, we find that photon noise has little impact for S/N >~ 10 per resolution element. However, the agreement between the autocorrelation along the line sight and the crosscorrelation between sightlines, a potential test of cosmological geometry, depends significantly on instrumental resolution. Finally, we present an inventory of metal lines. These include a a pair of strong C IV systems at z ~ 3.4 appearing only toward Q1439B, and a Mg II + Fe II system present toward Q1439 A and B at z = 1.68.

astro-ph↗

Supernova Remnants in the Fossil Starburst in M82

We report the discovery of ten compact H-alpha-bright sources in the post-starburst region northeast of the center of M82, ``M82 B.'' These objects have H alpha luminosities and sizes consistent with Type II supernova remnants (SNRs). They fall on the same H alpha surface brightness-diameter (Sigma-D) relation defined by SNRs in other nearby star-forming galaxies, with the M82 candidates lying preferentially at the small diameter end. These are the first candidates for optically-visible SNRs in M82 outside the heavily obscured central starburst within ~250 pc from the galactic center. If these sources are SNRs, they set an upper limit to the end of the starburst in region ``B2,'' about 500 pc from the galaxy's core, of ~50 Myr. Region ``B1,'' about 1000 pc from the core, lacks good SNR candidates and is evidently somewhat older. This suggests star formation in the galaxy has propagated inward toward the present-day intense starburst core.

astro-ph↗