Prof. Matt Jarvis

Abstract:

Forthcoming large photometric surveys for cosmology require precise and accurate photometric redshift (photo-z) measurements for the success of their main science objectives. However, to date, no method has been able to produce photo-zs at the required accuracy using only the broad-band photometry that those surveys will provide. An assessment of the strengths and weaknesses of current methods is a crucial step in the eventual development of an approach to meet this challenge. We report on the performance of 13 photometric redshift code single value redshift estimates and redshift probability distributions (PDZs) on a common set of data, focusing particularly on the 0.2pdbl-pdbl2.6 redshift range that the Euclid mission will probe. We designed a challenge using emulated Euclid data drawn from three photometric surveys of the COSMOS field. The data was divided into two samples: one calibration sample for which photometry and redshifts were provided to the participants; and the validation sample, containing only the photometry to ensure a blinded test of the methods. Participants were invited to provide a redshift single value estimate and a PDZ for each source in the validation sample, along with a rejection flag that indicates the sources they consider unfit for use in cosmological analyses. The performance of each method was assessed through a set of informative metrics, using cross-matched spectroscopic and highly-accurate photometric redshifts as the ground truth. We show that the rejection criteria set by participants are efficient in removing strong outliers, that is to say sources for which the photo-z deviates by more than 0.15(1pdbl+pdblz) from the spectroscopic-redshift (spec-z). We also show that, while all methods are able to provide reliable single value estimates, several machine-learning methods do not manage to produce useful PDZs. We find that no machine-learning method provides good results in the regions of galaxy color-space that are sparsely populated by spectroscopic-redshifts, for example zpdbl> pdbl1. However they generally perform better than template-fitting methods at low redshift (zpdbl< pdbl0.7), indicating that template-fitting methods do not use all of the information contained in the photometry. We introduce metrics that quantify both photo-z precision and completeness of the samples (post-rejection), since both contribute to the final figure of merit of the science goals of the survey (e.g., cosmic shear from Euclid). Template-fitting methods provide the best results in these metrics, but we show that a combination of template-fitting results and machine-learning results with rejection criteria can outperform any individual method. On this basis, we argue that further work in identifying how to best select between machine-learning and template-fitting approaches for each individual galaxy should be pursued as a priority.

Abstract:

Context: The LOFAR Two-metre Sky Survey (LoTSS) will eventually map the complete Northern sky and provide an excellent opportunity to study the distribution and evolution of the large-scale structure of the Universe.

Aims: We test the quality of LoTSS observations through a statistical comparison of the LoTSS first data release (DR1) catalogues to expectations from the established cosmological model of a statistically isotropic and homogeneous Universe.

Methods: We study the point-source completeness and define several quality cuts, in order to determine the count-in-cell statistics and differential source count statistics, and measure the angular two-point correlation function. We use the photometric redshift estimates, which are available for about half of the LoTSS-DR1 radio sources, to compare the clustering throughout the history of the Universe.

Results: For the masked LoTSS-DR1 value-added source catalogue, we find a point-source completeness of 99% above flux densities of 0.8 mJy. The counts-in-cell statistic reveals that the distribution of radio sources cannot be described by a spatial Poisson process. Instead, a good fit is provided by a compound Poisson distribution. The differential source counts are in good agreement with previous findings in deep fields at low radio frequencies and with simulated catalogues from the SKA Design Study and the Tiered Radio Extragalactic Continuum Simulation. Restricting the value added source catalogue to low-noise regions and applying a flux density threshold of 2 mJy provides our most reliable estimate of the angular two-point correlation. Based on the distribution of photometric redshifts and the Planck 2018 best-fit cosmological model, the theoretically predicted angular two-point correlation between 0.1 deg and 6 deg agrees reasonably well with the measured clustering for the sub-sample of radio sources with redshift information.

Conclusions: The deviation from a Poissonian distribution might be a consequence of the multi-component nature of a large number of resolved radio sources and/or of uncertainties on the flux density calibration. The angular two-point correlation function is < 10-2 at angular scales > 1 deg and up to the largest scales probed. At a 2 mJy flux density threshold and at a pivot angle of 1 deg, we find a clustering amplitude of A = (5.1? ±? 0.6) × 10-3 with a slope parameter of γ = 0.74? ±? 0.16. For smaller flux density thresholds, systematic issues are identified, which are most likely related to the flux density calibration of the individual pointings. We conclude that we find agreement with the expectation of large-scale statistical isotropy of the radio sky at the per cent level. The angular two-point correlation agrees well with the expectation of the cosmological standard model.

Abstract:

We present cosmological (30−400 Mpc) distributions of neutral hydrogen (H i) in the intergalactic medium (IGM) traced by Lyα emitters (LAEs) and QSOs at z = 2.1–2.5, selected with the data of the ongoing Hobby–Eberly Telescope Dark Energy Experiment (HETDEX) and the eBOSS survey. Motivated by a previous study of Mukae et al., we investigate spatial correlations of LAEs and QSOs with H i tomography maps reconstructed from H i Lyα forest absorption in the spectra of background galaxies and QSOs obtained by the CLAMATO survey and this study, respectively. In the cosmological volume far from QSOs, we find that LAEs reside in regions of strong H i absorption, i.e., H i rich, which is consistent with results of previous galaxy−background QSO pair studies. Moreover, there is an anisotropy in the H i distribution plot of transverse and line-of-sight distances; on average the H i absorption peak is blueshifted by ~200 km s−1 from the LAE Lyα redshift, reproducing the known average velocity offset between the Lyα emission redshift and the galaxy systemic redshift. We have identified a ~40 Mpc scale volume of H i underdensity that is a candidate for a giant H ii bubble, where six QSOs and an LAE overdensity exist at $\left\langle z\right\rangle =2.16$. The coincidence of the QSO and LAE overdensities with the H i underdensity indicates that the ionizing photon radiation of the QSOs has created a highly ionized volume of multiple proximity zones in a matter overdensity. Our results suggest an evolutionary picture where H i gas in an overdensity of galaxies becomes highly photoionized when QSOs emerge in the galaxies.

Abstract:

Context. Distant galaxy clusters provide an effective laboratory in which to study galaxy evolution in dense environments and at early cosmic times. Aims. We aim to identify distant galaxy clusters as extended X-ray sources that are coincident with overdensities of characteristically bright galaxies. Methods. We used optical and near-infrared data from the Hyper Suprime-Cam and VISTA Deep Extragalactic Observations (VIDEO) surveys to identify distant galaxy clusters as overdensities of bright, zphot = 0:8 galaxies associated with extended X-ray sources detected in the ultimate XMM extragalactic survey (XXL). Results. We identify a sample of 35 candidate clusters at 0:80 = z = 1:93 from an approximately 4.5 deg2 sky area. This sample includes 15 newly discovered candidate clusters, ten previously detected but unconfirmed clusters, and ten spectroscopically confirmed clusters. Although these clusters host galaxy populations that display a wide variety of quenching levels, they exhibit well-defined relations between quenching, cluster-centric distance, and galaxy luminosity. The brightest cluster galaxies (BCGs) within our sample display colours that are consistent with a bimodal population composed of an old and red sub-sample together with a bluer, more diverse sub-sample. Conclusions The relation between galaxy masses and quenching seem to already be in place at z ~ 1, although there is no significant variation in the quenching fraction with the cluster-centric radius. The BCG bimodality might be explained by the presence of a younger stellar component in some BCGs, but additional data are needed to confirm this scenario.

Abstract:

Many scientific investigations of photometric galaxy surveys require redshift estimates, whose uncertainty properties are best encapsulated by photometric redshift (photo-z) posterior probability density functions (PDFs). A plethora of photo-z PDF estimation methodologies abound, producing discrepant results with no consensus on a preferred approach. We present the results of a comprehensive experiment comparing 12 photo-z algorithms applied to mock data produced for The Rubin Observatory Legacy Survey of Space and Time Dark Energy Science Collaboration. By supplying perfect prior information, in the form of the complete template library and a representative training set as inputs to each code, we demonstrate the impact of the assumptions underlying each technique on the output photo-z PDFs. In the absence of a notion of true, unbiased photo-z PDFs, we evaluate and interpret multiple metrics of the ensemble properties of the derived photo-z PDFs as well as traditional reductions to photo-z point estimates. We report systematic biases and overall over/underbreadth of the photo-z PDFs of many popular codes, which may indicate avenues for improvement in the algorithms or implementations. Furthermore, we raise attention to the limitations of established metrics for assessing photo-z PDF accuracy; though we identify the conditional density estimate loss as a promising metric of photo-z PDF performance in the case where true redshifts are available but true photo-z PDFs are not, we emphasize the need for science-specific performance metrics.