Early Universe @UCL

This blog post was written by Boris Leistedt.

Devoured from within by supermassive black holes, quasars are among the most energetic and brightest objects in the universe. Their light sometimes travels several billion years before reaching us, and by looking at how they cluster in space, cosmologists are able to test models of the large-scale structure of the universe. However, being compact and distant objects, quasars look like stars and can only be definitively identified using high-resolution spectroscopic instruments. However, due to the time and expense of taking spectra, not all star-like objects can be examined with these instruments, and quasar candidates first need to be identified in a photometric survey and then confirmed or dismissed by taking follow-up spectra. This approach has led to the identification and study of tens of thousands of quasars, greatly enhancing our knowledge of the physics of these extreme objects.

Current catalogues of confirmed quasars are too small to study the large-scale structure of the universe at sufficient precision. For this reason, cosmologists use photometric catalogues of quasar candidates in which each object is only characterised by a small set of photometric colours. Star-quasar classification is difficult, yielding catalogues that in fact contain significant fractions of stars. In addition, the unavoidable variations in the calibration of instruments and in the observing conditions over time, create fluctuations in the number and properties of star-like objects detected on the sky. These observational issues combined with stellar contamination result in distortions in the data that can be misinterpreted as anomalies or hints of new physics.

In recent work we investigated these issues, and demonstrated techniques to address them. We considered the photometric quasars from the Sloan Digital Sky Survey (SDSS) and selected a subsample of objects where 95% of objects were expected to be actual quasars. We then constructed sky masks to remove the areas of the sky which were the most affected by calibration errors, fluctuations in the observing conditions, and dust in our own Galaxy. We exploited a technique called “mode projection” to obtain robust measurements of the clustering of quasars, and compared them with theoretical predictions. Using this, we found a remarkable agreement between the data and the prediction from the standard model of cosmology. Previous studies of such data argued that they were not suitable for cosmological studies, but we were able to identify a sample of objects that appear clean. In the future, we will use these techniques to analyse future photometric data, for example in the context of the Dark Energy Survey in which UCL is deeply involved.

Photometric quasars and some of their systematics

This guest blog post was written by Aurélien Benoit-Lévy.

The cosmic microwave background (CMB) is the furthest light we can observe. Since it is the furthest, bright galaxies and other compact objects can pollute in a nasty way the superb observations of the CMB that forthcoming experiments will soon deliver. To get rid of these point sources, we need to mask them. This is of course not harmless, as the resulting map is punched with holes just like emmental!

Holes in a map are unfortunate as they considerably complicate the spectral properties of the signal under consideration. But if there is a problem somewhere, it means that there is a solution, and I’m sure you would have guessed the solution: inpainting! Inpainting is just like cod, it’s good when it’s well made. The basic idea is to fill the holes with some fake data that would mimic some properties of the original signal. There are plenty of methods to inpaint a hole. For instance, the simplest one would be to fill the hole with the mean of all the surrounding pixels, but we can do much better than that.

CMB has a nice property: it’s correlated. This means that the probability that a given pixel has a given value is not independent of the value of the other pixels. And the CMB is also (very very close to) Gaussian, so we can use those two properties to generate random constrained Gaussian realizations that mimic the lost signal, but preserve the statistical structure of the map. Basically, we can use the information contained in the non-contaminated region to guess the most probable value in the holes. This technique has been known for quite a while, but sometimes requires heavy computations to accurately take into account all the information available in the map.

In recent work, my collaborators and I have demonstrated that it is not necessary to account for the whole map to inpaint a small hole. Using only the neighboring pixels is enough. With this method, we are able to inpaint a realistic CMB map, with a realistic point source mask in a realistic time! And the statistical properties of the inpainted maps are virtually unchanged! We applied this locally-constrained Gaussian realization in the special case of CMB lensing reconstruction and showed that this tool is very efficient, as it preserves the two-point statistic, and – above all – does not generate a spurious four-point signal. Of course, I now need to tell you about CMB lensing and four-point statistics, which is a very big thing and we’ll hear a lot about it in the coming months. But that will be for another post!

(a) CMB map with a point source hole, (b) neighbouring pixels used to inpaint the hole, (c) Input map, (d) Inpainted map

You can read more here:

Aurélien Benoit-Lévy et al.

Full-sky CMB lensing reconstruction in the presence of sky-cuts.

This guest blog post was written by Stephen Feeney.

Neutrinos are puzzling little particles. The Standard Model of particle physics tells us that there are three different types, or “flavours”, of neutrinos, each of which has no mass. Particle physics experiments, however, tell a different story. Firstly, the fact that neutrinos produced in the Sun (and nuclear reactors) are observed to oscillate between flavours requires them to have mass. Furthermore, unexpected results corroborated by a number of experiments measuring the rate of oscillations hint at the existence of more than three neutrino flavours.

This is all very interesting, and it seems that the Standard Model is ripe for revision, but what has it got to do with cosmology? Quite a lot, as it happens! Neutrinos, along with all fundamental particles, are produced in the early Universe. As they are so light, the neutrinos are extremely relativistic, and hence very energetic: so much so that, along with the photons, they dominate the dynamics of the early Universe. If there are four neutrinos rather than the standard three, the Universe initially expands more quickly than the current cosmological model (LambdaCDM) predicts, and cold dark matter becomes important at later times. Furthermore, neutrinos are so energetic that they are able to escape from the density perturbations set up in the early Universe — they “free-stream” a certain distance which is dependent on their mass — and hence affect the amount of structure that forms.

All of this has observable consequences for cosmology, affecting the relative amounts of simple elements produced in the Universe’s first few minutes, the amplitude and scale of the oscillations in the cosmic microwave background (CMB), and the numbers of, for example, galaxy clusters. We can therefore try to determine the number of neutrino species and their masses from cosmology! In fact, the strength of the effects of neutrinos in the early Universe, coupled with the quality of cosmological data, means that cosmological tests of neutrino physics are more powerful than laboratory experiments.

That’s what we (Hiranya, Licia Verde and I) have been doing. People have, of course, already started looking for hints of non-standard neutrinos in cosmology, and have found weak indications that the most likely number of neutrino species is greater than three. These analyses have, however, relied on parameter estimates: finding the most likely number of neutrino species given the data, and claiming evidence of new physics if this is not the standard value.

What we really want to do is to determine the support for a model with, for example, extra neutrinos, and compare this to the support for the standard cosmological model. This is what we’ve done: we have used the Bayesian evidence to compare the relative probabilities of LambdaCDM and models with extra neutrino flavours and massive neutrinos, given cosmological data. The datasets we consider include CMB power spectra from WMAP and the South Pole Telescope, measurements of the Universe’s recent expansion, and measurements of the weak gravitational lensing of the CMB by intervening large-scale structure. Our model-selection results indicate that the standard cosmological model is favoured over models with additional neutrinos and/or neutrino mass. There is therefore no requirement from cosmology to change our view of the neutrino. We’re going to need better cosmological data — such as the Planck data coming in a couple of months’ time — to make progress on this problem.

The lensing of the CMB by intervening matter: a promising cosmological probe of neutrino properties.

You can read more here:

Stephen M. Feeney, Hiranya V. Peiris and Licia Verde

Is there evidence for additional neutrino species from cosmology? JCAP, submitted, 2013.

Written by visiting collaborator and guest blogger Matt Johnson.

Computer simulations of the early universe allow us to explore scenarios that are just not possible to examine using pencil-and-paper calculations. I was recently in London working on how to use computer simulations to predict the observable signatures of eternal inflation. In the standard story of eternal inflation, our universe is purported to be enclosed inside one bubble among many, each expanding into an inflating background space-time. Our bubble undergoes collisions with others, providing a possible observational window on this bizarre scenario (read about the latest tests in this paper).

The field configuration making up our bubble determines the properties of our cosmology. For example, the symmetry of a single bubble fixes the universe to be an open Friedmann-Robertson-Walker (FRW) space-time. However, the collision between bubbles upsets this symmetry, meaning that if we live inside a bubble which underwent collisions, there will be deviations from a purely open FRW space-time.

The equations governing the evolution of the fields making up the bubbles and gravity are highly non-linear, making computer simulations a necessary tool for finding the exact space-time resulting from the collision between bubbles. Setting our sights on understanding the collision between pairs of bubbles, there is enough symmetry to evolve in 1 space and  1 time dimension. Last year, we wrote a code to simulate the collision between two bubble universes (you can read about our first results in this paper). One shortcoming of this original work was that the coordinates we used prevented us from evolving for longer than a few e-folds of inflation. We’ve fixed that by introducing a more appropriate set of coordinates, and now the challenge is to extract the cosmology contained inside one of the bubbles.

For colliding bubbles, one no longer has a purely homogeneous and isotropic cosmology. However, if collisions are to be consistent with what we observe, there had better be regions where the space-time is approximately FRW. So, how does one identify which regions are approximately FRW from the simulation data? How does one then quantify the deviation from FRW, so that quantitative predictions can be made? Our current method tracks the behaviour of geodesics evolved through the simulation. Because we know how geodesics would evolve through a purely FRW universe, we can determine the qualitative and quantitative differences. This is depicted schematically in the figure below, where I’ve shown the results of a simulation, and some of the geodesics evolving in the background. The geodesics entering the affected region pile up, indicating a deviation from pure FRW. We hope to polish the final code off soon, and extract the first observational signatures from our simulations!

The anatomy of a bubble collision