arXiv subjects

David Woods

Publications and source records attributed to David Woods.

At least 19 recordsLinked to original sources

arXiv2019

The WiggleZ Dark Energy Survey: Final Data Release and the Metallicity of UV-Luminous Galaxies

The WiggleZ Dark Energy Survey measured the redshifts of over 200,000 UV-selected (NUV<22.8 mag) galaxies on the Anglo-Australian Telescope. The survey detected the baryon acoustic oscillation signal in the large scale distribution of galaxies over the redshift range 0.2<z<1.0, confirming the acceleration of the expansion of the Universe and measuring the rate of structure growth within it. Here we present the final data release of the survey: a catalogue of 225415 galaxies and individual files of the galaxy spectra. We analyse the emission-line properties of these UV-luminous Lyman-break galaxies by stacking the spectra in bins of luminosity, redshift, and stellar mass. The most luminous (-25 mag < MFUV <-22 mag) galaxies have very broad H-beta emission from active nuclei, as well as a broad second component to the [OIII] (495.9 nm, 500.7 nm) doublet lines that is blue shifted by 100 km/s, indicating the presence of gas outflows in these galaxies. The composite spectra allow us to detect and measure the temperature-sensitive [OIII] (436.3 nm) line and obtain metallicities using the direct method. The metallicities of intermediate stellar mass (8.8<log(M*/Msun)<10) WiggleZ galaxies are consistent with normal emission-line galaxies at the same masses. In contrast, the metallicities of high stellar mass (10<log(M*/Msun)<12) WiggleZ galaxies are significantly lower than for normal emission-line galaxies at the same masses. This is not an effect of evolution as the metallicities do not vary with redshift; it is most likely a property specific to the extremely UV-luminous WiggleZ galaxies.

astro-ph.GA↗

arXiv2019

Bayesian optimal design for ordinary differential equation models with application in biological science

Bayesian optimal design is considered for experiments where the response distribution depends on the solution to a system of non-linear ordinary differential equations. The motivation is an experiment to estimate parameters in the equations governing the transport of amino acids through cell membranes in human placentas. Decision-theoretic Bayesian design of experiments for such nonlinear models is conceptually very attractive, allowing the formal incorporation of prior knowledge to overcome the parameter dependence of frequentist design and being less reliant on asymptotic approximations. However, the necessary approximation and maximization of the, typically analytically intractable, expected utility results in a computationally challenging problem. These issues are further exacerbated if the solution to the differential equations is not available in closed-form. This paper proposes a new combination of a probabilistic solution to the equations embedded within a Monte Carlo approximation to the expected utility with cyclic descent of a smooth approximation to find the optimal design. A novel precomputation algorithm reduces the computational burden, making the search for an optimal design feasible for bigger problems. The methods are demonstrated by finding new designs for a number of common models derived from differential equations, and by providing optimal designs for the placenta experiment.

stat.ME↗

arXiv2018

acebayes: An R Package for Bayesian Optimal Design of Experiments via Approximate Coordinate Exchange

We describe the R package acebayes and demonstrate its use to find Bayesian optimal experimental designs. A decision-theoretic approach is adopted, with the optimal design maximising an expected utility. Finding Bayesian optimal designs for realistic problems is challenging, as the expected utility is typically intractable and the design space may be high-dimensional. The package implements the approximate coordinate exchange algorithm to optimise (an approximation to) the expected utility via a sequence of conditional one-dimensional optimisation steps. At each step, a Gaussian process regression model is used to approximate, and subsequently optimise, the expected utility as the function of a single design coordinate (the value taken by one controllable variable for one run of the experiment). In addition to functions for bespoke design problems with user-defined utility functions, acebayes provides functions tailored to finding designs for common generalised linear and nonlinear models. The package provides a step-change in the complexity of problems that can be addressed, enabling designs to be found for much larger numbers of variables and runs than previously possible. We provide tutorials on the application of the methodology for four illustrative examples of varying complexity where designs are found for the goals of parameter estimation, model selection and prediction. These examples demonstrate previously unseen functionality of acebayes.

stat.CO↗

arXiv2018

Bayesian prediction for physical models with application to the optimization of the synthesis of pharmaceutical products using chemical kinetics

Quality control in industrial processes is increasingly making use of prior scientific knowledge, often encoded in physical models that require numerical approximation. Statistical prediction, and subsequent optimization, is key to ensuring the process output meets a specification target. However, the numerical expense of approximating the models poses computational challenges to the identification of combinations of the process factors where there is confidence in the quality of the response. Recent work in Bayesian computation and statistical approximation (emulation) of expensive computational models is exploited to develop a novel strategy for optimizing the posterior probability of a process meeting specification. The ensuing methodology is motivated by, and demonstrated on, a chemical synthesis process to manufacture a pharmaceutical product, within which an initial set of substances evolve according to chemical reactions, under certain process conditions, into a series of new substances. One of these substances is a target pharmaceutical product and two are unwanted by-products. The aim is to determine the combinations of process conditions and amounts of initial substances that maximize the probability of obtaining sufficient target pharmaceutical product whilst ensuring unwanted by-products do not exceed a given level. The relationship between the factors and amounts of substances of interest is theoretically described by the solution to a system of ordinary differential equations incorporating temperature dependence. Using data from a small experiment, it is shown how the methodology can approximate the multivariate posterior predictive distribution of the pharmaceutical target and by-products, and therefore identify suitable operating values. Materials to replicate the analysis can be found at www.github.com/amo105/chemicalkinetics.

stat.OT↗

arXiv2016

Measuring the 2D Baryon Acoustic Oscillation signal of galaxies in WiggleZ: Cosmological constraints

We present results from the 2D anisotropic Baryon Acoustic Oscillation (BAO) signal present in the final dataset from the WiggleZ Dark Energy Survey. We analyse the WiggleZ data in two ways: firstly using the full shape of the 2D correlation function and secondly focussing only on the position of the BAO peak in the reconstructed data set. When fitting for the full shape of the 2D correlation function we use a multipole expansion to compare with theory. When we use the reconstructed data we marginalise over the shape and just measure the position of the BAO peak, analysing the data in wedges separating the signal along the line of sight from that parallel to the line of sight. We verify our method with mock data and find the results to be free of bias or systematic offsets. We also redo the pre-reconstruction angle averaged (1D) WiggleZ BAO analysis with an improved covariance and present an updated result. The final results are presented in the form of $Ω_c h^2$, $H(z)$, and $D_A(z)$ for three redshift bins with effective redshifts $z = 0.44$, $0.60$, and $0.73$. Within these bins and methodologies, we recover constraints between 5% and 22% error. Our cosmological constraints are consistent with Flat $Λ$CDM cosmology and agree with results from the Baryon Oscillation Spectroscopic Survey (BOSS).

astro-ph.CO↗

arXiv2016

Bayesian Design of Experiments using Approximate Coordinate Exchange

The construction of decision-theoretic Bayesian designs for realistically-complex nonlinear models is computationally challenging, as it requires the optimization of analytically intractable expected utility functions over high-dimensional design spaces. We provide the most general solution to date for this problem through a novel approximate coordinate exchange algorithm. This methodology uses a Gaussian process emulator to approximate the expected utility as a function of a single design coordinate in a series of conditional optimization steps. It has flexibility to address problems for any choice of utility function and for a wide range of statistical models with different numbers of variables, numbers of runs and randomization restrictions. In contrast to existing approaches to Bayesian design, the method can find multi-variable designs in large numbers of runs without resorting to asymptotic approximations to the posterior distribution or expected utility. The methodology is demonstrated on a variety of challenging examples of practical importance, including design for pharmacokinetic models and design for mixed models with discrete data. For many of these models, Bayesian designs are not currently available. Comparisons are made to results from the literature, and to designs obtained from asymptotic approximations.

stat.ME↗

arXiv2016

GAMA/WiggleZ: The 1.4GHz radio luminosity functions of high- and low-excitation radio galaxies and their redshift evolution to z=0.75

We present radio Active Galactic Nuclei (AGN) luminosity functions over the redshift range 0.005 < z < 0.75. The sample from which the luminosity functions are constructed is an optical spectroscopic survey of radio galaxies, identified from matched Faint Images of the Radio Sky at Twenty-cm survey (FIRST) sources and Sloan Digital Sky Survey (SDSS) images.The radio AGN are separated into Low Excitation Radio Galaxies (LERGs) and High Excitation Radio Galaxies (HERGs) using the optical spectra. We derive radio luminosity functions for LERGs and HERGs separately in the three redshift bins (0.005 < z < 0.3, 0.3 < z < 0.5 and 0.5 < z <0.75). The radio luminosity functions can be well described by a double power-law. Assuming this double power-law shape the LERG population displays little or no evolution over this redshift range evolving as ~$(1+z)^{0.06}$ assuming pure density evolution or ~ $(1+z)^{0.46}$ assuming pure luminosity evolution. In contrast, the HERG population evolves more rapidly, best fitted by ~$(1+z)^{2.93}$ assuming a double power-law shape and pure density evolution. If a pure luminosity model is assumed the best fitting HERG evolution is parameterised by ~$(1+z)^{7.41}$. The characteristic break in the radio luminosity function occurs at a significantly higher power (~1 dex) for the HERG population in comparison to the LERGs. This is consistent with the two populations representing fundamentally different accretion modes.

astro-ph.GA↗

arXiv2015

Multivariate emulation of computer simulators: model selection and diagnostics with application to a humanitarian relief model

We present a common framework for Bayesian emulation methodologies for multivariate-output simulators, or computer models, that employ either parametric linear models or nonparametric Gaussian processes. Novel diagnostics suitable for multivariate covariance-separable emulators are developed and techniques to improve the adequacy of an emulator are discussed and implemented. A variety of emulators are compared for a humanitarian relief simulator, modelling aid missions to Sicily after a volcanic eruption and earthquake, and a sensitivity analysis is conducted to determine the sensitivity of the simulator output to changes in the input variables. The results from parametric and nonparametric emulators are compared in terms of prediction accuracy, uncertainty quantification and scientific interpretability.

stat.ME↗

arXiv2015

The WiggleZ Dark Energy Survey: Probing the Epoch of Radiation Domination using Large Scale Structure

We place the most robust constraint to date on the scale of the turnover in the cosmological matter power spectrum using data from the WiggleZ Dark Energy Survey. We find this feature to lie at a scale of $k_0=0.0160^{+0.0041}_{-0.0035}$ [h/Mpc] (68% confidence) for an effective redshift of 0.62 and obtain from this the first-ever turnover-derived distance and cosmology constraints: a measure of the cosmic distance-redshift relation in units of the horizon scale at the redshift of radiation-matter equality (r_H) of D_V(z=0.62)/r_H=18.3 (+6.3/-3.3) and, assuming a prior on the number of extra relativistic degrees of freedom $N_{eff}=3$, constraints on the matter density parameter $Ω_Mh^2=0.136^{+0.026}_{-0.052}$ and on the redshift of matter-radiation equality $z_{eq}=3274^{+631}_{-1260}$. All results are in excellent agreement with the predictions of standard LCDM models. Our constraints on the logarithmic slope of the power spectrum on scales larger than the turnover is bounded in the lower limit with values only as low as -1 allowed, with the prediction of standard LCDM models easily accommodated by our results. Lastly, we generate forecasts for the achievable precision of future surveys at constraining $k_0$, $Ω_Mh^2$, $z_{eq}$ and $N_{eff}$. We find that BOSS should substantially improve upon the WiggleZ turnover constraint, reaching a precision on $k_0$ of $\pm$9% (68% confidence), translating to precisions on $Ω_Mh^2$ and $z_{eq}$ of $\pm$10% (assuming a prior $N_{eff}=3$) and on $N_{eff}$ of (+78/-56)% (assuming a prior $Ω_Mh^2=0.135$). This is sufficient precision to sharpen the constraints on $N_{eff}$ from WMAP, particularly in its upper limit. For Euclid, we find corresponding attainable precisions on $(k_0, Ω_Mh^2, N_eff)$ of (3,4,+17/-21)%. This represents a precision approaching our forecasts for the Planck Surveyor.

astro-ph.CO↗

arXiv2014

The WiggleZ Dark Energy Survey: Improved Distance Measurements to z = 1 with Reconstruction of the Baryonic Acoustic Feature

We present significant improvements in cosmic distance measurements from the WiggleZ Dark Energy Survey, achieved by applying the reconstruction of the baryonic acoustic feature technique. We show using both data and simulations that the reconstruction technique can often be effective despite patchiness of the survey, significant edge effects and shot-noise. We investigate three redshift bins in the redshift range 0.2<$z$<1, and in all three find improvement after reconstruction in the detection of the baryonic acoustic feature and its usage as a standard ruler. We measure model independent distance measures $D_{\mathrm V}(r_{\mathrm s}^\mathrm{fid}/r_{\mathrm s})$ of 1716 $\pm$ 83 Mpc, 2221 $\pm$ 101 Mpc, 2516 $\pm$ 86 Mpc (68% CL) at effective redshifts z = 0.44, 0.6, 0.73, respectively, where $D_{\mathrm V}$ is the volume-average-distance, and $r_{\mathrm s}$ is the sound horizon at the end of the baryon drag epoch. These significantly improved 4.8, 4.5 and 3.4 percent accuracy measurements are equivalent to those expected from surveys with up to 2.5 times the volume of WiggleZ. These measurements are fully consistent with cosmologies allowed by the analyses of the Planck Collaboration and the Sloan Digital Sky Survey.We provide the $D_{\mathrm V}(r_{\mathrm s}^\mathrm{fid}/r_{\mathrm s})$ posterior probability distributions and their covariances. When combining these measurements with temperature fluctuations measurements of Planck, the polarization of WMAP9, and the 6dF Galaxy Survey baryonic acoustic feature, we do not detect deviations from a flat LCDM model. Assuming this model we constrain the current expansion rate to $H_0$ = 67.15 $\pm$ 0.98 kms$^{-1}$Mpc$^{-1}$. Allowing the equation of state of dark energy to vary we obtain $w_\mathrm{DE}$ = -1.080 $\pm$ 0.135. When assuming a curved LCDM model we obtain a curvature value of $Ω_{\mathrm K}$ = -0.0043 $\pm$ 0.0047.

astro-ph.CO↗

arXiv2013

The WiggleZ Dark Energy Survey: Star-formation in UV-luminous galaxies from their luminosity functions

We present the ultraviolet (UV) luminosity function of galaxies from the GALEX Medium Imaging Survey with measured spectroscopic redshifts from the first data release of the WiggleZ Dark Energy Survey. This sample selects galaxies with high star formation rates: at 0.6 < z < 0.9 the median star formation rate is at the upper 95th percentile of optically-selected (r<22.5) galaxies and the sample contains about 50 per cent of all NUV < 22.8, 0.6 < z < 0.9 starburst galaxies within the volume sampled. The most luminous galaxies in our sample (-21.0>M_NUV>-22.5) evolve very rapidly with a number density declining as (1+z)^{5\pm 1} from redshift z = 0.9 to z = 0.6. These starburst galaxies (M_NUV<-21 is approximately a star formation rate of 30 \msuny) contribute about 1 per cent of cosmic star formation over the redshift range z=0.6 to z=0.9. The star formation rate density of these very luminous galaxies evolves rapidly, as (1+z)^{4\pm 1}. Such a rapid evolution implies the majority of star formation in these large galaxies must have occurred before z = 0.9. We measure the UV luminosity function in 0.05 redshift intervals spanning 0.1<z<0.9, and provide analytic fits to the results. At all redshifts greater than z=0.55 we find that the bright end of the luminosity function is not well described by a pure Schechter function due to an excess of very luminous (M_NUV<-22) galaxies. These luminosity functions can be used to create a radial selection function for the WiggleZ survey or test models of galaxy formation and evolution. Here we test the AGN feedback model in Scannapieco et al. (2005), and find that this AGN feedback model requires AGN feedback efficiency to vary with one or more of the following: stellar mass, star formation rate and redshift.

astro-ph.CO↗

arXiv2013

The WiggleZ Dark Energy Survey: constraining galaxy bias and cosmic growth with 3-point correlation functions

Higher-order statistics are a useful and complementary tool for measuring the clustering of galaxies, containing information on the non-gaussian evolution and morphology of large-scale structure in the Universe. In this work we present measurements of the three-point correlation function (3PCF) for 187,000 galaxies in the WiggleZ spectroscopic galaxy survey. We explore the WiggleZ 3PCF scale and shape dependence at three different epochs z=0.35, 0.55 and 0.68, the highest redshifts where these measurements have been made to date. Using N-body simulations to predict the clustering of dark matter, we constrain the linear and non-linear bias parameters of WiggleZ galaxies with respect to dark matter, and marginalise over them to obtain constraints on sigma_8(z), the variance of perturbations on a scale of 8 Mpc/h and its evolution with redshift. These measurements of sigma_8(z), which have 10-20% accuracies, are consistent with the predictions of the LCDM concordance cosmology and test this model in a new way.

astro-ph.CO↗

arXiv2013

The WiggleZ Dark Energy Survey: measuring the cosmic growth rate with the two-point galaxy correlation function

The growth history of large-scale structure in the Universe is a powerful probe of the cosmological model, including the nature of dark energy. We study the growth rate of cosmic structure to redshift $z = 0.9$ using more than $162{,}000$ galaxy redshifts from the WiggleZ Dark Energy Survey. We divide the data into four redshift slices with effective redshifts $z = [0.2,0.4,0.6,0.76]$ and in each of the samples measure and model the 2-point galaxy correlation function in parallel and transverse directions to the line-of-sight. After simultaneously fitting for the galaxy bias factor we recover values for the cosmic growth rate which are consistent with our assumed $Λ$CDM input cosmological model, with an accuracy of around 20% in each redshift slice. We investigate the sensitivity of our results to the details of the assumed model and the range of physical scales fitted, making close comparison with a set of N-body simulations for calibration. Our measurements are consistent with an independent power-spectrum analysis of a similar dataset, demonstrating that the results are not driven by systematic errors. We determine the pairwise velocity dispersion of the sample in a non-parametric manner, showing that it systematically increases with decreasing redshift, and investigate the Alcock-Paczynski effects of changing the assumed fiducial model on the results. Our techniques should prove useful for current and future galaxy surveys mapping the growth rate of structure using the 2-dimensional correlation function.

astro-ph.CO↗

arXiv2013

The Stellar Masses of ~40,000 UV Selected Galaxies from the WiggleZ Survey at 0.3<z<1.0: Analogues of Lyman Break Galaxies?

We characterise the stellar masses and star formation rates in a sample of almost 40000 spectroscopically confirmed UV luminous galaxies at 0.3 5sigma in the UKIDSS-LAS at all redshifts. An even more luminous subset of 15% are also detected in the WISE 3.4 and 4.6um bands. We compute stellar masses for this very large sample of extremely blue galaxies and quantify the sensitivity of the stellar mass estimates to various assumptions made during the SED fitting. The median stellar masses are log10(M*/M0)=9.6\pm0.7, 10.2\pm0.5 and 10.4\pm0.4 for the IR-undetected, UKIDSS detected and UKIDSS+WISE detected galaxies respectively. We demonstrate that the inclusion of NIR photometry can lead to tighter constraints on the stellar masses. The mass estimates are found to be most sensitive to the inclusion of secondary bursts of star formation as well as changes in the stellar population synthesis models, both of which can lead to median discrepancies of the order of 0.3dex in the stellar masses. We find that the best-fit M/LK is significantly lower (by ~0.4 dex) than that predicted by simple optical colour based estimators, in particular for the bluer galaxies with younger best-fit ages. The WiggleZ galaxies have star formation rates of 3-10 M0/yr and mostly lie at the upper end of the main sequence of star-forming galaxies at these redshifts. Their rest-frame UV luminosities and stellar masses are comparable to both local compact UV-luminous galaxies as well as Lyman break galaxies at z~2-3.(abridged)

astro-ph.CO↗

arXiv2012

The WiggleZ Dark Energy Survey: Final data release and cosmological results

This paper presents cosmological results from the final data release of the WiggleZ Dark Energy Survey. We perform full analyses of different cosmological models using the WiggleZ power spectra measured at z=0.22, 0.41, 0.60, and 0.78, combined with other cosmological datasets. The limiting factor in this analysis is the theoretical modelling of the galaxy power spectrum, including non-linearities, galaxy bias, and redshift-space distortions. In this paper we assess several different methods for modelling the theoretical power spectrum, testing them against the Gigaparsec WiggleZ simulations (GiggleZ). We fit for a base set of 6 cosmological parameters, {Omega_b h^2, Omega_CDM h^2, H_0, tau, A_s, n_s}, and 5 supplementary parameters {n_run, r, w, Omega_k, sum m_nu}. In combination with the Cosmic Microwave Background (CMB), our results are consistent with the LambdaCDM concordance cosmology, with a measurement of the matter density of Omega_m =0.29 +/- 0.016 and amplitude of fluctuations sigma_8 = 0.825 +/- 0.017. Using WiggleZ data with CMB and other distance and matter power spectra data, we find no evidence for any of the extension parameters being inconsistent with their LambdaCDM model values. The power spectra data and theoretical modelling tools are available for use as a module for CosmoMC, which we here make publicly available at http://smp.uq.edu.au/wigglez-data . We also release the data and random catalogues used to construct the baryon acoustic oscillation correlation function.

astro-ph.CO↗

arXiv2012

The WiggleZ Dark Energy Survey: the transition to large-scale cosmic homogeneity

We have made the largest-volume measurement to date of the transition to large-scale homogeneity in the distribution of galaxies. We use the WiggleZ survey, a spectroscopic survey of over 200,000 blue galaxies in a cosmic volume of ~1 (Gpc/h)^3. A new method of defining the 'homogeneity scale' is presented, which is more robust than methods previously used in the literature, and which can be easily compared between different surveys. Due to the large cosmic depth of WiggleZ (up to z=1) we are able to make the first measurement of the transition to homogeneity over a range of cosmic epochs. The mean number of galaxies N(<r) in spheres of comoving radius r is proportional to r^3 within 1%, or equivalently the fractal dimension of the sample is within 1% of D_2=3, at radii larger than 71 \pm 8 Mpc/h at z~0.2, 70 \pm 5 Mpc/h at z~0.4, 81 \pm 5 Mpc/h at z~0.6, and 75 \pm 4 Mpc/h at z~0.8. We demonstrate the robustness of our results against selection function effects, using a LCDM N-body simulation and a suite of inhomogeneous fractal distributions. The results are in excellent agreement with both the LCDM N-body simulation and an analytical LCDM prediction. We can exclude a fractal distribution with fractal dimension below D_2=2.97 on scales from ~80 Mpc/h up to the largest scales probed by our measurement, ~300 Mpc/h, at 99.99% confidence.

astro-ph.CO↗

arXiv2012

The WiggleZ Dark Energy Survey: Joint measurements of the expansion and growth history at z < 1

We perform a joint determination of the distance-redshift relation and cosmic expansion rate at redshifts z = 0.44, 0.6 and 0.73 by combining measurements of the baryon acoustic peak and Alcock-Paczynski distortion from galaxy clustering in the WiggleZ Dark Energy Survey, using a large ensemble of mock catalogues to calculate the covariance between the measurements. We find that D_A(z) = (1205 +/- 114, 1380 +/- 95, 1534 +/- 107) Mpc and H(z) = (82.6 +/- 7.8, 87.9 +/- 6.1, 97.3 +/- 7.0) km/s/Mpc at these three redshifts. Further combining our results with other baryon acoustic oscillation and distant supernovae datasets, we use a Monte Carlo Markov Chain technique to determine the evolution of the Hubble parameter H(z) as a stepwise function in 9 redshift bins of width dz = 0.1, also marginalizing over the spatial curvature. Our measurements of H(z), which have precision better than 7% in most redshift bins, are consistent with the expansion history predicted by a cosmological-constant dark-energy model, in which the expansion rate accelerates at redshift z < 0.7.

astro-ph.CO↗

arXiv2012

The WiggleZ Dark Energy Survey: Cosmological neutrino mass constraint from blue high-redshift galaxies

The absolute neutrino mass scale is currently unknown, but can be constrained from cosmology. The WiggleZ high redshift star-forming blue galaxy sample is less sensitive to systematics from non-linear structure formation, redshift-space distortions and galaxy bias than previous surveys. We obtain a upper limit on the sum of neutrino masses of 0.60eV (95% confidence) for WiggleZ+Wilkinson Microwave Anisotropy Probe. Combining with priors on the Hubble Parameter and the baryon acoustic oscillation scale gives an upper limit of 0.29eV, which is the strongest neutrino mass constraint derived from spectroscopic galaxy redshift surveys.

astro-ph.CO↗