MPTA Finds Evidence of Gravitational Wave Background

An exciting suite of new scientific results have been published today, with early career researchers in radio astronomy from Australian institutions leading the international studies. The new results have found further evidence of the elusive gravitational wave background - the rattling and rolling of the fabric of the cosmos - measured by timing the ticks of nature’s extremely accurate and natural atomic clocks: millisecond pulsars.

In the first study, Dr. Matt Miles (Swinburne University of Technology / OzGrav) and his team utilised one of the world’s most sensitive telescopes - MeerKAT, an array made of 64 dish antennas spread out across the radio-quiet region of Karoo in South Africa. This study encapsulated carefully observing these millisecond pulsars over the course of 4.5 years, using MeerKAT’s L-band receiver (which translates to a frequency bandwidth of 856 - 1712 MHz).

Dr. Miles works as a member of the MeerKAT Pulsar Timing Array (MPTA) project, which - like other global pulsar timing array projects - is dedicated to using these rapidly rotating neutron stars to search for the gravitational wave backgrounds that permeate across our Universe. These experiments, of which there are now several around the world (Australia, USA, Europe, India, China, Argentina) utilise millisecond pulsars to look for these low-frequency gravitational waves.

“Studying the background lets us tune into the echoes of cosmic events across billions of years,” Dr. Miles explained. “It reveals how galaxies, and the universe itself, have evolved over time.”

In this latest work, Dr. Miles and his team have announced that they have now also uncovered further evidence of the gravitational wave background, thought to be generated by the merging of supermassive black hole binaries, across the Universe’s history. Their findings are published in the journal, MNRAS. These new results from the MPTA now provide an independent and supportive perspective to the results announced in June 2023 by the other PTA teams, in which the strongest evidence to date had been found of the gravitational wave background.

“What we’re seeing hints at a much more dynamic and active universe than we anticipated,” Dr Miles said. “We know supermassive black holes are out there merging, but now we’re starting to ask: where are they, and how many are out there?”

Gravitational waves (GWs) were predicted a little over 100 years ago, as a result of Einstein’s Theory of General Relativity. Einstein himself wrote the paper, but a few years later considered it to be a mistake. The world would have to wait until 1974 before the first indirect evidence of GWs would emerge, and then in 2015 when the first detection of GWs was announced.

GWs are caused by accelerating masses of any size - for example, a person doing squats will cause them. However, it is only the largest masses in our Universe, such as neutron stars and black holes, that cause ripples in space-time that are significant enough for human instruments to detect. These are still minuscule, the inaugural 2015 event caused variations to space-time on the scale of 1/10,000th the diameter of a proton - which is why we don’t feel them as they wash over our bodies, stretching and squeezing us right now.

Like the electromagnetic spectrum, GWs also come in a spectrum, with different masses producing different frequencies. Established ground-based detectors that use beams of lasers in giant tunnels that stretch for kilometres are tuned to detect GW events caused by stellar-mass black holes or merging neutron stars, usually having very short wave periods (seconds to minutes).

However, pulsar timing arrays have much longer baselines (extending many light-years) and so the focus of these experiments is on much lower-frequency GWs, which have wave periods that instead last decades from crest to crest. These nanohertz-regime GWs are thought to be formed by the supermassive black holes in the hearts of galaxies, as they inspiral around each other, eventually colliding and merging. 

The trouble is that the feeble signal from the gravitational wave background is buried within a lot of noise, and many different parameters need to be taken into account in these complex analyses. These parameters need to be understood to the best degree possible so that they can be de-coupled from the overall noise - leaving behind only the gravitational wave signal. This includes noise introduced from terrestrial issues, such as radio frequency interference, or the instrumentation within the telescope itself. Beyond this, the interstellar medium causes the smearing of each pulsar’s signal as it travels from the star to Earth. Even the pulsars themselves, whilst stable, have unique and intrinsic variations that need to be accounted for. For many parameters, often some assumptions need to be made, which can affect the overall outcome and results of these studies.

“The signals we receive from pulsars are incredibly predictable, down to nanosecond precision,” said Dr. Miles. “However, the combination of the pulse interacting with material in the space between the Earth and pulsars over thousands of light years, and the slight instabilities pulsars have as they rotate, result in delays that we need to understand.” 

“It’s like trying to pick out the tune of a particular instrument when listening to a band, it's a lot easier if you know what the other musicians are doing.”

However, once as much of this noise has all been quantified and modelled, the analysis and search for the gravitational wave background can commence. Since millisecond pulsars are extremely stable clocks located across many directions in the sky, astronomers like Dr. Miles can use radio telescopes like MeerKAT to observe these exotic objects over many years. These observations occur at a regular cadence, roughly once per month which then go into the analysis pipeline. Curious variations in one pulsar’s time series data might present some interesting science for that particular pulsar, but scientists are looking for a common signal amongst all millisecond pulsars.

The gravitational wave background signal is revealed in two manners -  the first is a common uncorrelated noise amongst all pulsars that shows consistent spectral properties. Scientists started noticing this signal appearing around 2021, however, studies revealed that this could be a mimic, red herring-type of event. The real smoking gun appears when the cross-correlated signal of all pulsars shows a particular pattern - known as the Hellings-Downs Curve - which highlights how the pulsars across all angles of the sky are affected. Evidence of this signal was found and announced in the June 2023 results, and these new findings from the MPTA results point towards the same outcomes.

The findings of the noise properties of the MPTA results have also been published in a second paper.

These new results, obtained independently by the MPTA team, were published about 1.5 years after the results that other PTAs published. MeerKAT is one of the most sensitive radio telescopes in the world and has been able to achieve this result in such a short timeframe (~4.5 years) because of this sensitivity. Other PTA teams, for example, the Parkes Pulsar Timing Array (PPTA) - which uses the Parkes radio telescope (Murriyang) have been running this experiment for over two decades to achieve similar outcomes (this is expected since Murriyang is a single-dish instrument).

"The combination of using a next-generation telescope and the international efforts behind developing new techniques to interrogate the data made this possible,” said Dr. Miles.

“MeerKAT is so sensitive that it can collect observations in under five minutes, which would take an hour for a telescope like Murriyang. This lets us look at many more pulsars over the years to greater precision than other experiments are able to, allowing us to get this result so quickly.”

When comparing the MPTA results with other PTAs, Dr Miles and his team found that their results were slightly different, with the largest difference between MPTA and PPTA and the closest match to MPTA being with the Chinese Pulsar Timing Array results.

“Broadly, the results are consistent across all of the experiments, which is great. However, it does look like the signal that we've measured is a little bit more powerful, which we don't expect should happen.” 

“Some of the results from other experiments have already indicated that the signal might be changing, and this could be an extension of that. This is incredibly exciting, and it opens up many more questions about what we might find in the future.”

Whilst these first pieces of evidence have indicated that we are only now starting to measure the gravitational wave background - a new view into a portion of the GW spectrum that has not been observed before - the science behind understanding it and its features still has a lot of work ahead. This translates to pulsar timing array campaigns continually observing their pulsar sets, and under the guise of the International Pulsar Timing Array project (a consortium of all the PTAs), combining their data to further improve sensitivities. This work is currently underway.

“We're just at the beginning of opening this new window to the Universe, and a lot of what we've found has led to many more questions to answer,” said Dr. Miles.

“The most important thing for us is to continue collecting data and to keep assessing the nature of this signal. More than that, we hope that active collaboration with the International Pulsar Timing Array will help us understand what we, and everyone else, are observing.”

One exciting area of interest is creating very detailed maps of gravitational waves across the Universe, as was the work of OzGrav / Monash University PhD candidate Rowina Nathan, a co-author on the third paper of this release. 

Nathan’s revealed an intriguing anomaly, an unexpected hotspot in the signal that could possibly suggest a directional bias. Read more from Nathan on these fascinating findings here.