# Publications

We present a numerical study of the evolution of galaxy clustering whengalaxies flow passively from high redshift, respecting the continuityequation throughout. While passive flow is a special case of galaxyevolution, it allows a well-defined study of galaxy ancestry and servesas an interesting limit to be compared to nonpassive cases. We usedissipationless N-body simulations, assign galaxies to massive halos atz=1 and 2 using various halo occupation distribution (HOD) models, andtrace these galaxy particles to lower redshift while conserving theirnumber. We find that passive flow results in an asymptotic convergenceat low redshift in the HOD and in galaxy clustering on scales above ~3h^{-1} Mpc for a wide range of initial HODs. As galaxies becomeless biased with respect to mass asymptotically with time, the HODparameters evolve such that M_{1}/M_{min} decreaseswhile α converges toward unity, whereg(M)>=exp(-M_{min}/M)[1+(M/M_{1})^{α}].The satellite populations converge toward the Poisson distribution atlow redshift. The convergence is robust for different number densitiesand is enhanced when galaxies evolve from higher redshift. We compareour results with the observed luminous red galaxy (LRG) sample from SDSSthat has the same number density. We claim that if LRGs have experienceda strict passive flow, their g(M)> should be close toa power law with an index of unity in halo mass. Discrepancies could bedue to dry galaxy merging or new members arising between the initial andthe final redshifts. The spatial distribution of passively flowinggalaxies within halos appears on average more concentrated than the halomass profile at low redshift. The evolution of bias for passivelyflowing galaxies is consistent with linear bias evolution onquasi-linear as well as large scales.

Many theoretical models of dark energy invoke rolling scaler fieldswhich in turn predict time varying values of the fundamental constants.Establishing the value of the fundamental constants at various times inthe universe can probe and test the various dark energy theories. One ofthe constants that is predicted to vary is the ratio of the electron toproton mass μ. It was established early on that molecular spectra aresensitive to the value of μ and can be used as probes of that value.This article describes the use of the spectrum of molecular hydrogen inhigh redshift Damped Lyman Alpha systems (DLAs) as a sensitive probe ofthe time evolution of μ.

We measure the evolution of the luminous red galaxy (LRG) luminosityfunction in the redshift range 0.12, minimizing the impact oflarge-scale structure on our results. We find that the LRG populationhas evolved little beyond the passive fading of its stellar populationssince z~0.9. Based on our luminosity function measurements and assuminga nonevolving Salpeter stellar initial mass function, we find that themost massive (L>3L^{*}) red galaxies have grown by less than

We present high signal-to-noise ratio measurements of the acoustic scalein the presence of nonlinear growth and redshift distortions using 320h^{-3} Gpc^{3} of cosmological particle-meshsimulations. Using simple fitting methods, we obtain robust measurementsof the acoustic scale with scatter close to that predicted by the Fishermatrix. We detect and quantify the shift in the acoustic scale byanalyzing the power spectrum: we detect at greater than 5 σ adecrease in the acoustic scale in the real-space matter power spectrum

Context: We have observed again two stars inside the ZZ Ceti instabilitystrip that were previously classified as not-observed-to-vary (NOV) byMukadam et al. (2004) and found them to be low-amplitude variables. Someevidence points to a pure ZZ Ceti instability strip; other evidencecontests it.

Aims: The two stars previously classified as NOVhave Sloan Digital Sky Survey (SDSS) spectroscopic effectivetemperatures that place them inside the ZZ Ceti instability strip, andthey were "contaminating" the strip as constant stars, which couldindicate that the instability strip was no longer a simple evolutionarystage. A pure instability strip indicates that pulsation is a normalphase which all DAs must go through.

Methods: We used effectivetemperatures derived from SDSS optical spectra by comparing them withmodel atmospheres to look for pulsators through time-resolved photometryand stars previously classified as NOV.

Results: Our new resultsindicate, but do not prove, a pure instability strip, because there arestill other NOV stars that need to be observed again. Additionally, wehave discovered five other ZZ Ceti stars based on their effectivetemperatures.Partially based on observations at Observatório do Pico dosDias/LNA; the Southern Astrophysical Research telescope, a collaborationbetween CNPq-Brazil, NOAO, UNC, and MSU; and McDonald Observatory of TheUniversity of Texas at Austin.

We describe the construction of MegaZ-LRG, a photometric redshiftcatalogue of over one million luminous red galaxies (LRGs) in theredshift range 0.4 < z < 0.7 with limiting magnitude i < 20.The catalogue is selected from the imaging data of the Sloan Digital SkySurvey (SDSS) Data Release 4. The 2dF-SDSS LRG and Quasar (2SLAQ)spectroscopic redshift catalogue of 13000 intermediate-redshift LRGsprovides a photometric redshift training set, allowing use of ANNz, aneural network-based photometric-redshift estimator. The rms photometricredshift accuracy obtained for an evaluation set selected from the 2SLAQsample is σ_{z} = 0.049 averaged over all galaxies, andσ_{z} = 0.040 for a brighter subsample (i < 19.0). Thecatalogue is expected to contain ~5 per cent stellar contamination. TheANNz code is used to compute a refined star/galaxy probability based ona range of photometric parameters; this allows the contaminationfraction to be reduced to 2 per cent with negligible loss of genuinegalaxies. The MegaZ-LRG catalogue is publicly available on the WorldWide Web from http://www.2slaq.info.

We measure the cosmological matter density by observing the positions ofbaryon acoustic oscillations in the clustering of galaxies in the SloanDigital Sky Survey (SDSS). We jointly analyze the main galaxies and LRGsin the SDSS DR5 sample, using over half a million galaxies in total. The

We present a Fourier analysis of the clustering of galaxies in thecombined main galaxy and LRG SDSS DR5 sample. The aim of our analysis isto consider how well we can measure the cosmological matter densityusing the signature of the horizon at matter-radiation equality embeddedin the large-scale power spectrum. The new data constrain the powerspectrum on scales 100-600 h^{-1} Mpc with significantly higherprecision than previous analyses of just the SDSS main galaxies, due toour larger sample and the inclusion of the LRGs. This improvement meansthat we can now reveal a discrepancy between the shape of the measuredpower and linear CDM models on scales 0.01 hMpc^{-1}-1, with linear model fitsfavoring a lower matter density (Ω_{M}=0.22+/-0.04) onscales 0.01 h Mpc^{-1}-1 and ahigher matter density (Ω_{M}=0.32+/-0.01) when smallerscales are included, assuming a flat ΛCDM model with h=0.73 andn_{s}=0.96. This discrepancy could be explained byscale-dependent bias, and by analyzing subsamples of galaxies, we findthat the ratio of small-scale to large-scale power increases with galaxyluminosity, so all of the SDSS galaxies cannot trace the same powerspectrum shape over 0.01 h Mpc^{-1}-1. However, the data are insufficient to clearly show aluminosity-dependent change in the largest scale at which a significantincrease in clustering is observed, although they do not rule out suchan effect. Significant scale-dependent galaxy bias on large scales,which changes with the r-band luminosity of the galaxies, couldpotentially explain differences in our Ω_{M} estimates anddifferences previously observed between 2dFGRS and SDSS power spectraand the resulting parameter constraints.

NICMOS observations of the resolved object fluxes in the Hubble DeepField-North and the Hubble Ultra Deep Field are significantly below thefluxes attributed to a 1.4-1.8 μm near-infrared background excess(NIRBE) from previous low spatial resolution NIRS measurements. Testsplacing sources in the NICMOS image with fluxes sufficient to accountfor the NIRBE indicate that the NIRBE flux must be either flat on scalesgreater than 100" or clumped on scales of several arcminutes to avoiddetection in the NICMOS image. A fluctuation analysis of the new NICMOSdata shows a fluctuation spectrum consistent with that found at the samewavelength in deep 2MASS calibration images. The fluctuation analysisshows that the majority of the fluctuation power comes from resolvedgalaxies at redshifts of 1.5 and less and that the fluctuations observedin the earlier deep 2MASS observations can be completely accounted forwith normal low-redshift galaxies. Neither the NICMOS direct fluxmeasurements nor the fluctuation analysis require an additionalcomponent of near-infrared flux other than the flux from normal resolvedgalaxies in the redshift range between 0 and 7. The residualfluctuations in the angular range between 1" and 10" is 1-2 nWm^{-2} sr^{-1}, which is at or above several predictionsof fluctuations from high redshift Population III objects, butinconsistent with attributing the entire NIRBE to high-redshiftgalaxies.

We have discovered 21 new Type Ia supernovae (SNe Ia) with the HubbleSpace Telescope (HST) and have used them to trace the history of cosmicexpansion over the last 10 billion yr. These objects, which include 13spectroscopically confirmed SNe Ia at z>=1, were discovered during 14epochs of reimaging of the GOODS fields North and South over 2 yr withthe Advanced Camera for Surveys on HST. Together with a recalibrationof our previous HST-discovered SNe Ia, the full sample of 23 SNe Ia atz>=1 provides the highest redshift sample known. Combining these datawith previous SN Ia data sets, we measured H(z) at discrete,

We present a new estimator, ω, of the small-scale galaxycorrelation function that is robust against the effects ofredshift-space distortions and large-scale structures. The estimator isa weighted integral of the redshift space or angular correlationfunction and is a convolution of the real-space correlation functionwith a localized filter. This allows a direct comparison with theory,without modelling redshift-space distortions and the large-scalecorrelation function. This has a number of advantages over the moretraditional w_{p} estimator, including (i) an insensitivity tolarge-scale structures and the details of the truncation of theline-of-sight integral, (ii) a well-localized kernel in ξ(r) and(iii) being unbinned. We discuss how this estimator would be used inpractice, applying it to a sample of mock galaxies selected from theMillennium simulation.

An optical/infrared telescope of 20-100m aperture located on the Moonwould be able to observe objects 100 to 1,000 times fainter than theproposed next generation of space telescopes. The infrared region of thespectrum is particularly important for observations of objects atredshifts z>7. The apparent simplicity and low mass of a liquidmirror telescope, compared with a traditional pointable glass mirror,suggest that the concept should be considered further. A previouslyproposed liquid mirror telescope, based upon a spinning liquid metallicalloy, is not appropriate for infrared applications, which will requirea liquid below 130K. Here we report the successful coating of an ionicliquid with silver. The surface is smooth and the silver coating isstable on a timescale of months. The underlying ionic liquid does notevaporate in a vacuum and remains liquid down to a temperature of 175K.Given that there are ~10^{6} simple and ~10^{18} ternaryionic liquids, it should be possible to synthesize liquids with evenlower melting temperatures.

We present the 3D real-space clustering power spectrum of a sample of~600000 luminous red galaxies measured by the Sloan Digital Sky Survey,using photometric redshifts. These galaxies are old, elliptical systemswith strong 4000-Å breaks, and have accurate photometric redshiftswith an average error of Δz = 0.03. This sample of galaxies rangesfrom redshift z = 0.2 to 0.6 over 3528 deg^{2} of the sky,probing a volume of 1.5h^{-3}Gpc^{3}, making it thelargest volume ever used for galaxy clustering measurements. We measurethe angular clustering power spectrum in eight redshift slices and usewell-calibrated redshift distributions to combine these into ahigh-precision 3D real-space power spectrum from k = 0.005 to k =1hMpc^{-1}. We detect power on gigaparsec scales, beyond theturnover in the matter power spectrum, at a ~2σ significance for k< 0.01hMpc^{-1}, increasing to 5.5σ for k <0.02hMpc^{-1}. This detection of power is on scalessignificantly larger than those accessible to current spectroscopicredshift surveys. We also find evidence for baryonic oscillations, bothin the power spectrum, as well as in fits to the baryon density, at a2.5 σ confidence level. The large volume and resulting smallstatistical errors on the power spectrum allow us to constrain both theamplitude and the scale dependence of the galaxy bias in cosmologicalfits. The statistical power of these data to constrain cosmology is ~1.7times better than previous clustering analyses. Varying the matterdensity and baryon fraction, we find Ω_{M} = 0.30 +/-0.03, and Ω_{b}/Ω_{M} = 0.18 +/- 0.04, for afixed Hubble constant of 70kms^{-1}Mpc^{-1} and ascale-invariant spectrum of initial perturbations. The detection ofbaryonic oscillations also allows us to measure the comoving distance toz = 0.5; we find a best-fitting distance of 1.73 +/- 0.12Gpc,corresponding to a 6.5 per cent error on the distance. These resultsdemonstrate the ability to make precise clustering measurements withphotometric surveys.

We discuss the effects of nonlinear structure formation on the signatureof acoustic oscillations in the late-time galaxy distribution. We arguethat the dominant nonlinear effect is the differential motion of pairsof tracers separated by 150 Mpc. These motions are driven by bulk flowsand cluster formation and are much smaller than the acoustic scaleitself. We present a model for the nonlinear evolution based on thedistribution of pairwise Lagrangian displacements that provides aquantitative model for the degradation of the acoustic signature, evenfor biased tracers in redshift space. The Lagrangian displacementdistribution can be calibrated with a significantly smaller set ofsimulations than would be needed to construct a precise power spectrum.By connecting the acoustic signature in the Fourier basis with that inthe configuration basis, we show that the acoustic signature is morerobust than the usual Fourier-space intuition would suggest, because thebeat frequency between the peaks and troughs of the acousticoscillations is a very small wavenumber that is well inside the linearregime. We argue that any possible shift of the acoustic scale is

We present measurements of the redshift-space three-point correlationfunction of 50967 luminous red galaxies (LRGs) from Data Release 3 (DR3)of the Sloan Digital Sky Survey (SDSS). We have studied the shapedependence of the reduced three-point correlation function(Q_{z}(s, q, θ)) on three different scales, s = 4, 7 and10h^{-1}Mpc, and over the range of 1 < q < 3 and 0°< θ < 180°. On small scales (s = 4h^{-1}Mpc),Q_{z} is nearly constant, with little change as a function of qand θ. However, there is evidence for a shallow U-shapedbehaviour (with θ) which is expected from theoretical modellingof Q_{z}(s, q, θ). On larger scales (s = 7 and10h^{-1}Mpc), the U-shaped anisotropy in Q_{z} (withθ) is more clearly detected. We compare this shape dependence inQ_{z}(s, q, θ) with that seen in mock galaxy catalogueswhich were generated by populating the dark matter haloes in largeN-body simulations with mock galaxies using various halo occupationdistributions (HOD). We find that the combination of the observed numberdensity of LRGs, the (redshift-space) two-point correlation function andQ_{z}(s, q, θ) provides a strong constraint on theallowed HOD parameters (M_{min}, M_{1},α) andbreaks key degeneracies between these parameters. For example, ourobserved Q_{z}(s, q, θ) disfavours mock catalogues thatoverpopulate massive dark matter haloes with many LRG satellites. Wealso estimate the linear bias of LRGs to be b = 1.87 +/- 0.07 inexcellent agreement with other measurements.

The baryon acoustic oscillations are a promising route to the precisionmeasure of the cosmological distance scale and hence the measurement ofthe time evolution of dark energy. We show that the nonlineardegradation of the acoustic signature in the correlations oflow-redshift galaxies is a correctable process. By suitablereconstruction of the linear density field, one can sharpen the acousticpeak in the correlation function or, equivalently, restore the higherharmonics of the oscillations in the power spectrum. With this, one canachieve better measurements of the acoustic scale for a given surveyvolume. Reconstruction is particularly effective at low redshift, wherethe nonlinearities are worse but where the dark energy density ishighest. At z=0.3, we find that one can reduce the sample variance errorbar on the acoustic scale by at least a factor of 2 and in principle bynearly a factor of 4. We discuss the significant implications ourresults have for the design of galaxy surveys aimed at measuring thedistance scale through the acoustic peak.

A Spitzer Space Telescope survey in the NOAO Deep Wide Field inBoötes provides a complete, 8 μm-selected sample of galaxies toa limiting (Vega) magnitude of 13.5. In the 6.88 deg^{2} field

We present the cosmological distance errors achievable using thebaryonic acoustic oscillations as a standard ruler. We begin from aFisher matrix formalism that is upgraded from earlier Seo and Eisensteinwork. We isolate the information from the baryonic peaks by excludingdistance information from other less robust sources. Meanwhile, weaccommodate the Lagrangian displacement distribution into the Fishermatrix calculation to reflect the gradual loss of information in scaleand in time due to nonlinear growth, nonlinear bias, and nonlinearredshift distortions. We then show that we can contract themultidimensional Fisher matrix calculations into a two-dimensional oreven one-dimensional formalism with physically motivated approximations.We present the resulting fitting formula for the cosmological distanceerrors from galaxy redshift surveys as a function of survey parametersand nonlinearity, which saves us going through the 12 dimensional Fishermatrix calculations. Finally, we show excellent agreement between thedistance error estimates from the revised Fisher matrix and theprecision on the distance scale recovered from N-body simulations.

This paper extends our previous fluctuation analysis of thenear-infrared background at 1.6 μm to the 1.1 μm (F110W) image ofthe Hubble Ultra Deep Field. When all detectable sources are removed,the ratio of fluctuation power in the two images is consistent with theratio expected for faint, z<8, sources, and is inconsistent with theexpected ratio for galaxies with z>8. We also use numericallyredshifted model galaxy spectral energy distributions for 50 and 10 Myrold galaxies to predict the expected fluctuation power at 3.6 and 4.5μm to compare with recent Spitzer observations. The predictedfluctuation power for galaxies at z=0-12 matches the observed Spitzerfluctuation power, while the predicted power for z>13 galaxies ismuch higher than the observed values. As was found in the 1.6 μm(F160W) analysis, the fluctuation power in the source-subtracted F110Wimage is 2 orders of magnitude below the power in the image with allsources present. This leads to the conclusion that the 0.8-1.8 μmnear-infrared background is due to resolved galaxies in the redshiftrange z<8, with the majority of power in the redshift range of0.5-1.5.

We present preliminary results of an ESO-VLT large programme (AMAZE)aimed at determining the evolution of the mass-metallicity relation atz~3 by means of deep near-IR spectroscopy. Gas metallicities and stellarmasses are measured for an initial sample of nine star forming galaxiesat z~3.3. When compared with previous surveys, the mass-metallicityrelation inferred at z~3.3 shows an evolution significantly strongerthan observed at lower redshifts. There are also some indications thatthe metallicity evolution of low mass galaxies is stronger relative tohigh mass systems, an effect which can be considered as the chemicalversion of the galaxy downsizing. The mass-metallicity relation observedat z~3.3 is difficult to reconcile with the predictions of somehierarchical evolutionary models. We shortly discuss the possibleimplications of such discrepancies.