MONDAY, 23 SEPTEMBER 2024
13.8 billion years of our universe’s history have been investigated in the past century by scientists looking at the sky. From the outside, it looks like theorists write models and equations to describe the Universe, astronomers point their telescopes at the sky – ‘magic’ happens - and eventually they compare findings to f ind the best model. This is pretty much what we call cosmology.The standard model of cosmology encompasses numerous quantities that cannot be directly measured, such as the mass of a hypothetical particle, or expansion velocity. There are lots of equations to extract the quantity from the ones measured. For example, to know more about the birth of the Universe, you need to match all models between that primordial time and today’s observations. Some people do this as a job, connecting blackboards and telescopes. Usually, the more numerous the steps, the more difficult it is to precisely estimate a quantity or confirm a whole theory. Many cosmologists are pioneering statisticians. However, despite their great talent, the availability of good data – in quantity, quality, and diversity - a common language is still crucial. This common language is a quantity which is directly observable: spacetime correlation.
THE POWER OF CORRELATION
Correlation, a concept used in many fields, describes a relationship between two objects, whether qualitative or quantitative. For a risk analyst, as much as for a journalist, claiming correlation between observations is the first step towards identifying causes and effects. The correlation has a mathematical definition: given repeated measures of two or more events, it calculates the similarity of their variations, for example at different times. However, “correlation is not causation” is drilled into us all in science class. Therefore, it does not mean that any of those observed phenomena is the source of the others; there might just be an external factor linking them. Sometimes this does not matter. After all, a trader only needs to know which markets are correlated to avoid putting all their eggs in the same basket. This hidden factor matters for a cosmologist because it usually means that something is sourcing all those events. Something lying in the past, closer to the birth of the Universe.
Let’s take an example on Earth. One can think of waves rolling towards the beach and take snapshots of random points. The sea level at each point is different, but across all the snapshots, each point’s level will have a high correlation in space between sea levels at specific points. Correlated points are separated by the same distance, called the wavelength. A wave physicist could predict this wavelength with a model of the waves’ source: wind and tidal effects. In modelling, parameters are adjusted based on observations, akin to measuring ocean wave wavelengths to reveal missing factors- the force of the wind. Models failing to align with data are ruled out, leaving ‘standard models’ that best match observations. Modelling the universe is like navigating a sea of particles, fluids and astrophysical objects, and seeking the optimal model tuned by diverse observational data.
A SKY FULL OF…
Unlike the sea, the source of data for cosmologists and astronomers is the sky. Observations are primarily based on three types of signals: light, particles and more recently, gravitational waves, thanks to telescopes, interacting detectors and interferometers respectively, either from space or the Earth. So far, most discoveries about the Universe have used light observations coming from stars, galaxies, supernovae (imploded old stars), clusters of galaxies and more. Astrophysicists can identify them based on their size and redshift. Redshift is the type of light and from how far it was stretched (to redder frequencies) when travelling to us. If one sees light as a wave, an expanding universe stretches the wavelength more the more it travels. The position of the object and its redshift tell us where and when it comes from: we can calculate a spacetime correlation of the sky.
THE STORY OF THE STORY: PRE-BIG BANG PHYSICS
Since 1929, we have known that the Universe is expanding, so even light takes longer to travel between clusters of galaxies. Going back in time, the universe becomes hotter and denser, presenting opportunities for interactions and initiating correlations. In the 1940s it was realised that the Universe must have once been a ‘soup’ of constantly scattering light and elementary particles before diluting by expanding. It was predicted that the free light emitted from this young Universe would still be travelling to us today. Indeed, in 1963 the ‘cosmic microwave background’ photons (light particles) were detected all over the sky. This helped tune the standard model for the expansion of the Universe and how its contents changed between the ‘soup’ era and today.
Correlation comes in for the next discovery. Those photons have tiny differences in energy, as if the sky was a sea of very small waves. Computing the correlation between any two points’ energy gives a surprising result. Photons very far apart show correlated waves, even those too far apart to have ever been in contact. It seemed impossible, as if photons were travelling along a too-fast expanding road, unable to catch up with those in front, and yet somehow now showing evidence of having crashed into each other. In the 1980s, this called for a pre-Big Bang era in which those particles would be able to interact: this is the theory of inflation. There are hundreds of inflationary sub-models because we don’t know what the Universe was made of at that time, just that it expanded unequivocally fast. It is a general framework with some principles and a lot to tune: the structure of the Universe, the particles it has and the forces which exist.
These models assume that the earliest interactions were at small enough scales to be quantum. Quantum physics is relevant at small scales, except that here both particle physics and gravity are at play. Probing the validity of this is thus about testing ‘theories of everything’. Despite this, correlations are still helpful, and even fundamental, because they can provide a complete description of complex systems of quantum particles. When the Universe becomes big enough, those are usual spacetime correlations with typical signatures for us to spot.
FUTURE DISCOVERIES
Many models were ruled out but infinitely many remain. How can the inflation model be tested then? Inflation makes unprecedented predictions. First, all models predict that the cosmic microwave background is slightly non-Gaussian. In other words, waves of different wavelengths interacted and created new ones with special signatures. This implies that you need to know how much three points or more correlate, not just two by two. Experiments have detected this already, but the effect is too small compared to our precision to be able to confirm. Inflationary theory also predicts the existence of primordial gravitational waves, also sourced by the youngest quantum waves of the Universe. For now, they remain undetected, which still rules out many models. The next stage of experiments, such as the Simons Observatory and the CMB-S4 experiment, may bring a decisive test of inflation - go check them out! In 1969 everyone was excited about the first men on the Moon, but there are plenty of giant leaps for mankind still to come.
Article by Yoann Launay
Artwork by Josh Langfield
