1,721,431 research outputs found
Thinking outside the box; fast error estimation for next generation galaxy surveys
No full textSince the discovery of the accelerated expansion of the universe in the late 90s, the flat CDM model has reigned as the best explanation for the cosmological phenomena we have observed. In spite of over two decades of study, the identities and properties of both cold dark matter (CDM) and dark energy () remain a mystery. The goal of modern precision cosmology is to measure cosmological parameters using a multitude of probes in the pursuit of deviations from the CDM framework or insights into the properties of dark matter and dark energy. If the probe being considered is the matter power spectrum measured from a galaxy redshift survey, then the precision of the derived parameters is determined by the power spectrum covariance matrix. The analytic form of the covariance matrix is difficult to estimate, so common practice is to run many simulations of the survey volume to get a brute-force estimate.
Next-generation cosmological surveys are set to collect higher resolution data within a larger survey volume than ever before. The complexity and number of simulations that will be required to estimate the covariance matrix of these surveys is threatening to become too computationally expensive for even the most advanced computer clusters. Thus, there is an urgent need to develop novel techniques for reducing the computation time required to achieve such precise covariance estimates. While many proposed methods seek to reduce the number of simulations required, it is also possible to leverage the volume scaling of the covariance matrix, allowing one to reduce the size of the simulations required.
Super-sample covariance (SSC) is a contribution to the covariance matrix made by modes of the power spectrum that are larger than the volume of a survey or simulation. If this volume scaling of the covariance is to be taken advantage of, then the SSC within the simulations must be accurately modeled. To this end, I review methods of running separate universe (SU) simulations to account for the effects of SSC. While these methods have all been shown to recover the SSC with reasonable accuracy, they have been largely developed and tested in isolation from one another. I present my work in directly comparing the accuracy of these methods in recovering the SSC effect using ensembles of N-body simulations.
Even with SSC accurately modeled, the volume scaling of the covariance does not hold for arbitrarily small volume simulations; at some point, the analytic behaviour of the covariance is expected to break down. I push the volume scaling to its limit by running many thousands of simulations at different volumes and scaling the covariance to match that of a larger volume survey. The SSC term has a nontrivial relation to the simulation volume, preventing it from scaling in the same way as the other components of the covariance. In light of this, I present a way to include SSC such that the scaled covariance could still be recovered with good accuracy. I find the scaled covariance matches the large volume covariance to within or better on most scales, with higher bins being biased low due to missing a small component of the SSC. The scaled covariance at very low for very small simulations is substantially lower than the large mock covariance at those scales due to very few modes of that scale being present in the small volume simulations. This creates a skewness in the distribution of power at those scales. By computing the number of modes required to avoid this skewed distribution of power, I derive a way to estimate the minimum simulation volume that could be used to accurately model the covariance at a given scale. The accurate modeling of SSC and optimal leveraging of the volume scaling of the covariance matrix are powerful complementary tools with the potential to substantially reduce the computational cost of covariance matrix estimation for future galaxy survey data
An Exploration of Clustering Statistics and Tidal Alignments in the Context of Group Catalogues
Full text linkThe use of three-dimensional galaxy surveys and large cosmological simulations of the galaxy distribution has advanced our understanding of large-scale structure over the last few decades. At the forefront of this are clustering measurements that seek to determine cosmological parameters including the linear growth function, . When making clustering measurements using group catalogues, whether they be dark matter halos or galaxy clusters, various selections in our catalogues are generally performed that can alter the outputs for our measurements. An example of this is when objects are selected based on a certain mass threshold, a primary bias is introduced that alters our perceived two-point clustering statistics. Additionally, when using spectroscopic data to measure the positions of groups in the Universe, an observational effect caused by peculiar velocities, known as redshift space distortions, can also alter our perceived clustering and create an anisotropic effect in the power spectrum. Furthermore, many recent works have focused on secondary properties, independent of mass, that can alter clustering measurements for groups. In this work, we will explore selection-based contamination effects when finding groups in redshift space with a focus on intrinsic alignments for our selections. The selections used in this work consider halo size as a proxy for its mass and could lead to intrinsic alignments altering the anisotropic signal. We will utilize a current model that includes selection-based effects to the standard redshift space distortion model where this contamination incorporates a dependence on the configuration of the large-scale tidal fields and thus intrinsic alignments. If the selection of our groups is further biased, this can result in systematic errors to , altering growth of structure measurements. To investigate, we primarily use CDM simulations for dark matter only particles and explore how the selection of groups with preferential halo alignments can produce an anisotropic signal, even in real space. We will also create a variety of mock galaxy catalogues using a halo occupation distribution for our dark matter groups. This population technique is optimized and modified to include similar intrinsic alignment statistics as our dark matter halo catalogue. We will then perform various tests of intrinsic alignments on these galaxy catalogues and attempt to differentiate between anisotropic effects induced by such alignments in our selection with respect to a catalogue that lacks their presence. We find that utilizing various selection-based algorithms produced a weak differential anisotropic signal between these catalogues, most likely as a consequence of significant noise sources. These noise sources will be discussed and potential advancements to the limitations faced are considered for future work
Novel Techniques for the Calibration of Systematics in Next Generation Galaxy Surveys
No full textBaryon Acoustic Oscillation (BAO) observations offer a robust method for measuring cosmological expansion. However, the BAO signal in a sample of galaxies can be diluted and shifted by interlopers - galaxies that have been assigned the wrong redshifts. Because of the slitless spectroscopic method adopted by the Roman and Euclid space telescopes, the galaxy samples resulting from single line detections will have relatively high fractions of interloper galaxies. Interlopers with a small displacement between true and false redshift have the strongest effect on the measured clustering. In order to model the BAO signal, the fraction of such interlopers and their clustering need to be accurately known.
We introduce a new method to self-calibrate these quantities by shifting the contaminated sample towards or away from us along the line of sight by the interloper offset, and measuring the cross-correlations between these shifted samples. The contributions from the different components are shifted in scale in this cross-correlation compared to the auto-correlation of the contaminated sample, enabling the decomposition and extraction of the component terms. We demonstrate the application of the method using numerical simulations and show that an unbiased BAO measurement can be extracted. Unlike previous attempts to model the effects of contaminants, self-calibration allows us to make fewer assumptions about the form of the contaminants such as their bias.
We also introduce a new statistical technique to cosmology, called the Leave One-Out Probability Integral Transform (LOO-PIT), as a complementary test to the standard best
fit statistic χ2. This technique combines two concepts: LOO-CV (Leave One Out-Cross Validation), and the well known Probability Integral Transform (PIT). LOO-PIT primarily has the advantage of diagnosing the type of modelling failure as well as relaxing the constraint of assuming Gaussian likelihoods in one’s data analysis, paving the way for more general methods. While it is a general method, we apply LOO-PIT to the problem of diagnosing unknown interlopers in galaxy catalogues
Investigating future galaxy surveys of cosmological large-scale structure
Full text linkMy thesis consists of the two following projects:
1-Testing Large-Scale Structure Measurements against Fisher Matrix Analysis (Chapter 2):
We compare Baryonic Acoustic Oscillation (BAO) and Redshift Space Distortion (RSD) measurements from recent galaxy surveys with their Fisher matrix based predictions. Measurements of the position of the BAO signal lead to constraints on the comoving angular diameter distance and the Hubble distance that agree well with their Fisher matrix based expectations. However, RSD-based measurements of the growth rate do not agree with the predictions made before the surveys were undertaken, even when repeating those predictions using the actual survey parameters. We show that this is due to a combination of effects including degeneracies with the geometric parameters and , and optimistic assumptions about the scale to which the linear signal can be extracted. We show that measurements using current data and large-scale modelling techniques extract an equivalent amount of signal to that in the linear regime for , remarkably independent of the sample properties and redshifts covered.
2- Correcting for small-displacement interlopers in BAO analyses (Chapter 3):
Due to the low resolution of slitless spectroscopy, future surveys including those made possible by the Roman and Euclid space telescopes will be prone to line mis-identification, leading to interloper galaxies at the wrong redshifts in the large-scale structure catalogues. The most pernicious of these have a small displacement between true and false redshift such that the interloper positions are correlated with the target galaxies. We consider how to correct for such contaminants, focusing on interlopers in [O{\sc\,ii}] catalogues as will be observed by Roman, which are misplaced by at redshift . Because this displacement is close to the BAO scale, the peak in the interloper-target galaxy cross-correlation function at the displacement scale can change the shape of the BAO peak in the auto-correlation of the contaminated catalog, and lead to incorrect cosmological measurements if not accounted for properly. We consider how to build a model for the monopole and quadrupole moments of the contaminated correlation function, including an additional free parameter for the fraction of interlopers. The key input to this model is the cross-correlation between the population of galaxies forming the interlopers and the main target sample. It will be important to either estimate this using calibration data or to use the contaminated small-scale auto-correlation function to model it, which may be possible if a number of requirements about the galaxy populations are met. We find that this method is successful in measuring the BAO dilation parameters without significant degradation in accuracy provided the cross-correlation function is accurately known
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Precision Cosmology Using Voids
Full text linkIn the late 1990s, the discovery that the expansion rate of the Universe was accelerating was a decisive moment for cosmology. The last 25 years have seen the consolidation of this component, called dark energy, which dominates the total energy of the Universe at the present day. Cosmologists have developed many techniques to measure the properties of dark energy and attempt to reveal insights into the physics behind this mysterious component. Many explanations for dark energy exist, the simplest being that it is a form of energy permeating all of space (a cosmological constant), and alternatives include modifications to theories of general relativity and scalar fields. The modern era of precision cosmology has been dedicated to the measurement of cosmological parameters that describe and distinguish different models. Despite decades of work in this area, little insight has been found into the nature of dark energy. More accurate measurements from next-generation cosmological surveys are needed to uncover the underlying physics behind this fundamental component of the Universe.
Cosmic voids are patches of the Universe that are less dense than the cosmic average. These large-scale underdensities are a natural consequence of structure growth. Voids are special places in the Universe where the physics of their growth can be easily modelled. Although the density is non-linear (the density in the centre of voids is close to zero), the motions of galaxies still track their primordial form making it possible to extract cosmological information. This information primarily comes from two physical processes - the Alcock-Paczynski (AP) effect and Redshift-Space Distortions (RSDs). The AP effect is a geometrical consequence where an object's shape becomes distorted if measured using a wrong cosmological model. Stacking voids will produce a spherical averaged shape only if the AP parameter, D_M/D_H, is correct (where D_M is the transverse comoving distance that is a measure perpendicular to the line of sight and D_H is the Hubble distance which is a measure parallel to the line of sight). RSDs are the distortions of measured distances due to the Doppler effect of a galaxy's peculiar velocity. On large scales, the growth rate of cosmological structure is the dominant source of RSDs. Using the linear motions of galaxies around voids, RSDs are used to measure the growth rate of structure parameterised by f(z) sigma_8(z) (where f(z) relates to the growth rate of structure and sigma_8 relates to the redshift space galaxy power spectrum).
Measurements of voids within the large-scale structure of the Universe can be made using galaxy spectroscopic surveys. These surveys use the positions of galaxies as tracers of the underlying matter distribution. Information in these surveys has primarily been extracted using two techniques: Baryonic Acoustic Oscillations (BAO) and RSD. BAO provide a standard ruler through which the expansion rate of the Universe can be measured, while RSD allows for a measurement of the growth rate of structure. The use of voids has emerged as another technique to extract even more information from these surveys. This thesis presents the background and modelling that can be used to extract and analyze this information.
After all necessary background is summarised, measurements of the anisotropic cross-correlation of galaxies and cosmic voids in data from the Sloan Digital Sky Survey Main Galaxy Sample (MGS), Baryon Oscillation Spectroscopic Survey (BOSS) and extended BOSS (eBOSS) luminous red galaxy catalogues from SDSS Data Releases 7, 12 and 16, covering the redshift range 0.07<z<1.0 are presented. This uses the clustering of galaxies around voids to extract information and is the first time that a consistent analysis method has been applied to extract information from voids in this full redshift range. A reconstruction method is applied to the galaxy data before void-finding to remove selection biases when constructing the void samples. Results of a joint fit to the multipole moments of the measured cross-correlation for the growth rate of structure and the ratio D_M/D_H are reported in six redshift bins. For D_M/D_H, voids are able to achieve significantly higher precision than that obtained from analyses of BAO and RSD in the same datasets. Our growth rate measurements are of lower precision but still comparable with galaxy clustering results. For both quantities, the results agree well with the expectations for a LambdaCDM model. The degeneracy directions obtained for the study of voids in galaxy spectroscopic surveys are consistent with and complementary to those from other cosmological probes and result in a significant gain of information. These results consolidate void-galaxy cross-correlation measurements as a pillar of modern observational cosmology.
Also presented are cosmological models fits to voids and the combination of voids with other probes. A standard LambdaCDM cosmological model is fit to measurements from voids as well as various extensions including a constant dark energy equation of state not equal to -1, a time-varying dark energy equation of state, and these same models allowing for spatial curvature. Results on key parameters of these models are reported for void-galaxy and galaxy-galaxy clustering alone, both of these combined, and all these combined with measurements from the cosmic microwave background (CMB) and supernovae (SN). The results show a remarkable agreement with a flat LambdaCDM cosmology for all cosmological models tested. The gain of information from void measurements made at multiple redshifts, compared to compressing all information into one measurement at a single effective redshift, is also demonstrated.
Finally, a forward look to the future of voids as cosmological probes is presented. This thesis uses the best public galaxy redshift survey data available to date; however, this will soon be surpassed once DESI and Euclid results are released within the next few years. Forecast constraints from applying a consistent analysis method to that presented in this thesis on a mock catalogue expected to match data from Euclid are shown.
Cosmic voids provide another analysis method that can extract independent cosmological constraints with complementary parameter degeneracies that, combined with information from BAO/RSD, increase the precision of information extracted from galaxy spectroscopic surveys. Future surveys will need to continue to build on the current modelling of voids to reduce systematic errors and provide valuable hints towards the fundamental nature of our Universe
Optimizing Small-scale Redshift Space Distortion Measurements in eBOSS for Cosmological Inference
Full text linkModern cosmological datasets have grown substantially in size and the precision of their measurements. While the improvement has had a beneficial impact on our understanding of the cosmological model, it requires equal improvements in our analysis methods and the treatment of systematic biases to achieve optimal results. The model that best fits current observations is a spatially flat model with a cold dark matter (CDM) component that dominates the matter density and a cosmological constant (Λ) that dominates the energy density (ΛCDM). The objective of most cosmological datasets is to precisely measure the parameters of the model, discover an extension, or identify a tension with the expectations from another probe, with the eventual goal of discovering new physics. A probe of particular interest for this thesis is measurements of Redshift Space Distortions (RSD), which constrain the growth of structure through the parameter combination fσ₈, consisting of the logarithmic growth rate of density perturbations, f, and the amplitude of density fluctuations normalized using the standard deviation of fluctuations in a sphere of radius 8 Mpc/h, defined as σ₈. Not only do these measurements constrain core parameters of the ΛCDM model, they are also particularly interesting because they come from the velocity field rather than the density field directly. This makes them complementary to many other large-scale probes, and particularly useful for constraining theories of modified gravity.
One of the new datasets is the extended Baryon Oscillation Spectroscopic Survey (eBOSS), which spectroscopically observed over 1 million galaxies between 2014-2019 as part of the Sloan Digital Sky Survey (SDSS). I present the full eBOSS pipeline, from survey design to the cosmological analysis of the final data release (DR 16), with a focus on my contributions to its development. A key element is the treatment of observational systematics, which must be removed from the data to obtain reliable cosmological results. One of the most significant systematics for small-scale measurements is fibre collisions, where an observational limitation prevents observing close pairs of targets, producing a biased clustering measurement. I present the work of myself and my collaborators within eBOSS to generate Pairwise-Inverse-Probability (PIP) weights and combine them with Angular Upweighting (ANG) to fully remove the effect of fibre collisions, obtaining unbiased clustering measurements on all scales. I also describe my work to correct an observational systematic in the eBOSS Emission Line Galaxy (ELG) sample, caused by inconsistent calibration in the surveys used to identify targets for eBOSS, using a weight-based correction that does not require discarding already observed data. From the final cosmological analyses I present measurements of the Baryon Acoustic Oscillation (BAO) scale and RSD signal from each eBOSS sample. These measurements constrain the expansion history and growth history over the range 0.6 < z < 2.2, finding good agreement with the expectation from the 2018 Planck Cosmic Microwave Background (CMB) data for a flat ΛCDM model. When combined with other SDSS BAO measurements, as well as CMB and supernovae observations, we obtain precise measurements of the curvature of the Universe and the equation-of-state of the dark matter component, two of the simplest possible extensions to the cosmological model, and find both to be in agreement with flat ΛCDM to high precision.
Constraints on fσ₈ using small-scale RSD measurements have a significant statistical advantage over those made on large scales. My collaborators and I measure the small-scale clustering of the DR 16 eBOSS Luminous Red Galaxy (LRG) sample, using the PIP+ANG weights to correct for fibre collisions. We fit to the monopole and quadrupole moments of the 3D correlation function and to the projected correlation function over the separation range 7-60 Mpc/h with a model based on the AEMULUS cosmological emulator to measure fσ₈. We obtain a measurement of fσ₈(z=0.737)=0.408+/-0.038, which is 1.4σ lower than the value expected from Planck 2018 measurements for a flat ΛCDM model, and is more consistent with recent weak-lensing measurements. The level of precision achieved is 1.7 times better than more standard measurements made using only the large-scale modes of the same sample. We also fit to the data using the full range of scales modelled by the AEMULUS cosmological emulator, 0.1-60 Mpc/h, and find a 4.5σ tension in the amplitude of the halo velocity field with the Planck+ΛCDM model, driven by a mismatch on the non-linear scales. We perform a robust analysis of possible sources of systematics, including the effects of redshift uncertainty and incompleteness due to target selection that were not included in previous analyses fitting to clustering measurements on small scales.
The restriction of constraining fσ₈ using only the measurement scales 7-60 Mpc/h was motivated by the minimum scale at which the velocity scaling parameter used in the emulator to replicate changes in the growth rate still matched the expectation for a change in fσ₈. This issue highlights an important concern for small-scale RSD measurements: the need to carefully disentangle the linear and non-linear information when interpreting RSD in terms of fσ₈. It is particularly important to do this given the significant deviation from the expectation based on the Planck+ΛCDM model derived using the full range of scales modelled by the emulator in the previous analysis. We construct a new emulator-based model for small-scale galaxy clustering with scaling parameters for both the linear and non-linear velocities of galaxies, allowing us to isolate the linear growth rate. We train the emulator using simulations from the AbacusCosmos suite, estimating the linear velocity of galaxies by evolving the velocities of the simulations' Zel'dovich approximation initial conditions using linear growth. We apply a tophat smoothing kernel of radius 5 Mpc/h to the field to remove the remaining small-scale velocity dispersion, finding good agreement between the behaviour of our linear velocity scaling parameter and the expectation for a change in fσ₈ on all scales. We apply the new emulator to the eBOSS LRG sample, obtaining a value of fσ₈(z=0.737)=0.368+/-0.041, in 2.3σ tension with the Planck+ΛCDM expectation. We also find less dependence on the minimum measurement scale than the previous analysis, validating our improved emulator.
The small- and large-scale eBOSS results provide a precise test of ΛCDM from both the expansion and growth history. While consistent with ΛCDM, these measurements give interesting insight into the current H₀ and S₈ tensions between various cosmological probes, and give some evidence for a third tension between the fσ₈ measurements of small-scale RSD analyses and the Planck 2018+ΛCDM expectation. The observations and analysis of the eBOSS samples, particularly the treatment of observational systematics, pave the way for the next generation of surveys, such as those currently being done by the Dark Energy Spectroscopic Instrument (DESI) and Euclid space mission. Applying the small-scale RSD analysis method to these surveys will be critical to achieving optimal constraints, which have the potential to revolutionize the ΛCDM model and our understanding of the Universe
- …
