The Cosmic Messengers


By George Seabroke

Astronomy is no longer limited to light. It can now also now “see” cosmic events via particles (neutrinos and cosmic rays) and gravitational waves from distant objects. These “messengers” are opening up exciting new windows on the Universe. Multi-messenger astronomy is observing events through these new multiple windows close in time to learn even more about the event.

When you look at a star or galaxy, light from the object enters your eye, allowing you to detect it. Any neutrinos from the object will pass straight through your eye, your head and the Earth and you will be none the wiser! Coming the other way, there are 65 billion neutrinos per square centimetre per second from the Sun, entering your eye through the Earth and back of your head! Gravitational waves will also pass through your eye. If you are in space and close your eyes, you will be able to see cosmic rays flash in your eyes. Like the telescope enabling our eyes to collect more light, technology is now allowing us to “see” more of the reality that surrounds us, similar to how Neo, in The Matrix, learns to see the underlying components of his reality that are invisible to others.

The electromagnetic spectrum

Light is an electromagnetic wave and so has an associated wavelength (the distance over which the wave’s shape repeats). Fig. 1 illustrates the electromagnetic spectrum. The human eye has evolved to see optical light, which has wavelengths between 400 and 700 nanometres (one billionth of a metre). Like neutrinos and gravitational waves, most of the electromagnetic spectrum is invisible to the human eye! Our Sun’s light intensity is highest in the optical part of the electromagnetic spectrum and Earth’s atmosphere lets most of this optical light down to the ground. This is why the human eye has evolved to see this part of the electromagnetic spectrum. The atmosphere blocks the longest wavelength radio waves (longer than 10 metres) and the shortest wavelength gamma rays (shorter than 0.1 nanometres), as well as many other segments of the full light spectrum owing to a variety of molecules which are present.

10_03Like the human eye, the first telescope in 1608 collected optical light. As technology improved and additional types of radiation were discovered, telescopes evolved to observe radio and infrared light, which is transmitted through the atmosphere to the ground. The ozone layer blocks most of the emitted ultraviolet light, X-rays and gamma rays so observations at these wavelengths had to wait for the space age.

The longer the wavelength, the cooler the temperature of the phenomena. The Universe is full of cold, hydrogen gas that does not emit optical light. It does, however, emit light with 21 centimetre wavelengths, highlighting regions rife with active ionisation and star-formation processes. This allows ground-based radio telescopes (see Fig. 1) to map spiral galaxies and their interactions with other galaxies.

The remnant glow of the Big Bang, called the Cosmic Microwave Background, can still be seen in space in ubiquitous microwave photons, which have wavelengths between the radio and infrared. Its discovery led to the 1978 Nobel Prize for Physics. Fig. 1 includes the Wilkinson Microwave Anisotropy Probe (WMAP) satellite in space (finished in 2010).

Far-infrared observations can be made from the stratosphere. Fig. 1 shows the Stratospheric Observatory for Infrared Astronomy (SOFIA) aeroplane flying in the bottom left of the infrared region (still operational). Objects that emit lots of far-infrared photons are cooler than 140 K. These include the interstellar medium (gas and dust clouds between stars), star-forming regions, planetary atmospheres and surfaces, and comets.

Although some infrared light makes it to the Earth’s surface, the atmosphere itself radiates in the infrared, just like we do. Therefore, to observe the faintest objects at these wavelengths, it is best to go to space. This is exactly what the James Webb Space Telescope (JWST) will do in 2021 (it is already shown in space in Fig. 1, on the dividing line between infrared and optical regions). As light travels in our expanding Universe, its wavelength becomes red-shifted. Thus, the first galaxies forming just a few hundred million years after the Big Bang will be easier to see if viewed in the infrared, where not only the majority of their light is located, but where interstellar dust does a poor job of absorbing radiation.

Optical light makes it to the ground but being in space avoids atmospheric distortions. This allows star-mapping satellites, like the one I work on, called Gaia, to measure the tiny motions of stars extremely precisely over time. This is also why JWST’s predecessor, the Hubble Space Telescope, is in space and still going strong since its launch in 1990. Hubble is depicted in space in Fig. 1, on the dividing line between the optical and ultraviolet regions. This allows Hubble to find the most distant and oldest objects in the Universe. Currently, this is GN-z11, which is observed, as it existed 13.4 billion years ago, just 400 million years after the Big Bang. Hubble also observes in the ultraviolet. Hot objects produce ultraviolet photons, where the majority of ultraviolet light is produced by stars in the early and late stages of their evolution.

The X-ray satellite portrayed in Fig. 1 is the Chandra X-ray Observatory, which is still operational. Gas at temperatures from about a million kelvin to hundreds of millions of kelvin emit X-ray photons. Strong gravitational fields around neutron stars and black holes cause matter to accelerate and collide, heating the gas to these temperatures. This discovery was recognised in half of the 2002 Nobel Prize for Physics. Neutron stars are the collapsed core of stars, which, before they explode as supernova, had masses between about 8 and 40 times the mass of our Sun. When even more massive stars go supernova, their cores can form stellar mass black holes, from which not even light can escape.

The gamma-ray satellite portrayed in Fig. 1 is the Fermi Gamma-ray Space Telescope, which is still operational. Gamma-ray photons are produced in extreme events like supernovae, neutron star mergers, and blazars. Blazars are a particular type of Active Galactic Nuclei (AGN). AGN are powered by matter falling onto a supermassive black hole at the centre of the host galaxy, producing more light than expected from a galaxy of just stars. This also causes a relativistic jet (where matter is accelerated to close to the speed of light). If this jet is pointed at Earth, then the object is both an AGN and a blazar, since it “blazes” down your line-of-sight.

Therefore, we have a large suite of astronomical observatories, taking advantage of the full breadth of the electromagnetic spectrum. This allows astronomers to get a complete, multi-faceted picture of astronomical objects. For example, if we observe a galaxy in optical light only, we see its stars, star clusters and the dust lanes which happen to be silhouetted against background stars. If we limited ourselves to optical light only, we would miss dust lanes not illuminated from behind (emitting infrared light), hot young stars (emitting ultraviolet light), neutron stars (emitting X-ray light) and supernovae (which emit gamma rays). This means we would miss out on the beginning and end of the stellar life cycle, the ability to map the full structure of galaxies, and the capability of inferring how they were created and evolve.

Cosmic rays

The term cosmic ray actually refers to particles that have mass, as opposed to light rays (photons) that do not have mass. About 1% of cosmic rays are electrons. About 99% are the nuclei of atoms, stripped of their electrons. Of these, about 90% are protons (hydrogen atoms missing their single electron), 9% are helium nuclei and 1% are the nuclei of heavier elements.

The solar wind is composed of charged particles, with similar compositions as cosmic rays, which stream away from the Sun. The Sun’s upper atmosphere, the corona, accelerates these particles to 5000-10,000 electron volts (1 electron volt = 1.6 × 10-19 joules of energy). Flares close to the Sun’s lower atmosphere, the photosphere, or Coronal Mass Ejections in interplanetary space, can accelerate the solar wind particles up to much higher energies: tens of millions to tens of billions of electron volts. These particles are now classified as solar cosmic rays.

Non-solar cosmic rays are historically called Galactic cosmic rays, although some have their origins outside our Galaxy, making them extragalactic cosmic rays. The discovery of galactic cosmic rays, in 1912, led to the 1936 Nobel Prize for Physics. They primarily have energies similar to solar cosmic rays, but a small fraction possess energies that are much higher. Ultra-high-energy cosmic rays have been observed to have energies of 1020 electron volts: 40 million times more energetic than what the Large Hadron Collider can produce!

I encounter both types of cosmic rays in my work because they cause spurious flux in the detectors that I work with on the Gaia satellite. Unfortunately these hits do not reveal anything about their origin so they are “noise” that needs to be removed to allow Gaia to measure the positions and spectra of stars in our Galaxy. For other astronomers working on other missions, however, cosmic rays are exactly the signal they are trying to measure, and they use very different apparatuses to learn about their energies, locations, compositions and origins.

When a high-energy gamma-ray photon or cosmic ray hits the Earth’s atmosphere, it can create an “air shower” of secondary particles (which includes electrons). These particles are traveling in the air with a speed greater than that at which light propagates in air. Nature permits this because these speeds are still slower than the ultimate cosmic speed limit: the speed of light in a vacuum. Whenever particles move faster than light through a medium, they emit light called Cherenkov radiation, which is why the water tanks around nuclear reactors glow an eerie, blue colour. As the air shower moves through the atmosphere, it also causes the nitrogen present to fluoresce (emit ultraviolet light). Ground-based experiments such as the Major Atmospheric Gamma Imaging Cherenkov Telescope (MAGIC) in the Canary Islands use both types of light to determine the direction and energy of the original cosmic ray or gamma ray that struck Earth.

Large water tanks, and the ensuing Cherenkov radiation produced within them, can also be used to detect the secondary particles that reach the Earth. The Pierre Auger Observatory in Argentina combines both methods: water tanks (it has 1660 water tanks, each containing 12 tonnes of water and each 1.5 km apart) and telescopes. The latter can measure showers in more detail than the former but can only observe fluorescence on dark nights. Because the water tanks are active all the time, they measure 10 times more events. Using both methods together creates a more powerful instrument, allowing them to find that the arrival directions of the highest-energy cosmic rays are correlated with the positions of nearby AGN.


Like electrons, neutrinos are elementary particles but their mass is thought to be less than one-millionth the mass of the electron! Neutrinos were first detected in a nuclear reactor in 1956, which led to half of the 1995 Nobel Prize for Physics. Neutrinos from the Sun were first detected in 1968, which led to the other half of the aforementioned 2002 Nobel Prize for Physics.

Solar neutrinos are produced in the Sun’s core by nuclear fusion. This process is also the origin of all the Sun’s photons. Photons take hundreds of thousands of years to make it to the Sun’s surface because they scatter off protons on the way out. For this reason, photons are not direct measurements of the core. Neutrinos, on the other hand, interact with matter very rarely. Upon their creation, they fly out of the Sun’s core unimpeded and so are direct measurements of the core.

Neutrinos are notoriously difficult to detect because their interaction cross-section (the probability of interacting with another particle) is so small. Even lead shielding one light-year thick (9.5 trillion km or 6 trillion miles) would only stop half of the neutrinos flying through it!

Similar to detecting cosmic ray secondary particles, neutrinos also need large tanks of water for detection. The Kamiokande-II experiment had 3000 tonnes of water in the Kamioka mine, 1 km underground in Japan. At the time of its construction, the most recent supernova that had been observed in our Galaxy was in 1604. Therefore, it was fortuitous that Kamiokande-II started observing in 1985 and a supernova went off in 1987 (called 1987A) in our neighbouring galaxy, the Large Magellanic Cloud.

As the star’s core collapsed, approximately 10% of the star’s mass was converted into a ten-second burst of neutrinos, numbering 1058! Kamiokande-II detected 12 of these (two other experiments detected 8 and 5). As of 2018, it is still the only verified detection of neutrinos from a supernova, marking the beginning of extra-solar neutrino astronomy. These neutrinos arrived two or three hours before light from 1987A reached Earth. It was the first time neutrinos and light had been observed nearly simultaneously from a supernova, ushering in the era of multi-messenger astronomy, correlating electromagnetic (light-based) astronomy with neutrino (particle-based) astronomy.

Currently, the largest neutrino detector in the world is the IceCube Neutrino Observatory. It has thousands of sensors distributed within a cubic kilometre of ice (1 billion tonnes) in Antarctica! On 22nd September 2017, IceCube recorded an extremely high-energy neutrino. Within one minute of its detection, the IceCube Collaboration sent out an automated alert. This contained the co-ordinates of where the neutrino had come from for astronomers around the world to search for a possible source.

Searches of the co-ordinates (in the constellation of Orion) produced only one likely source: TXS 0506+056. This was a previously known blazar. Space observations by the Fermi satellite (28th September) and ground observations by MAGIC (4th October) detected gamma rays from the blazar. The nearly simultaneous detection of gamma rays and neutrinos from the same source can be explained by ultra-high-energy cosmic rays (protons) being accelerated in the blazer jets. This acceleration produces two different types of exotic particles, which decay into gamma rays and neutrinos. Two “messengers” were directly detected and used to indirectly infer the presence of a third messenger. This result suggests that blazars may be one of the sources of ultra-high-energy cosmic rays and thus the origin of a sizeable fraction of the cosmic neutrino flux observed by IceCube.

Gravitational waves

Gravitational waves are the disturbance of the fabric of spacetime (reality requires the three dimensions of space and the one dimension of time to be part of a four-dimensional continuum). Gravitational waves are generated by masses being accelerated. They propagate in all directions away from their source at the speed of light as ripples in spacetime.

Albert Einstein predicted them in 1916 but never believed they would be detected because of their tiny effect on the objects they pass through. Their effect is equivalent to changing the distance to the nearest star to the Sun, Proxima Centauri, by one hair’s width! Gravitational waves were indirectly inferred from the observations of a binary pulsar (spinning neutron star), which led to the 1993 Nobel Prize for Physics.

However, this is another example of Einstein not always being correct, as the first gravitational wave was directly measured in 2015. Its story began over a billion years ago when two black holes with 36 and 29 solar masses merged and emitted a series of gravitational waves. (At this time on Earth, the first sexually reproducing organisms were just evolving). Then, in the last 50 years of the gravitational wave’s journey to Earth, laser interferometry was established.

Laser interferometry, as used in our best modern gravitational wave detectors like LIGO and Virgo, is when a single laser beam is split and sent down to identical tubes at right angles to each other. Mirrors at either end of each tube reflect the light back to the detector. Undisturbed, the laser beams arrive simultaneously at the detector, where the phases of the two beams are perfectly aligned. When those two laser beams are recombined, they cancel each other out (destructive interference), and no laser light is detected.

If the laser beams are disturbed by a gravitational wave, however, one of the tubes will contract and the other will expand. This is because gravitational waves alternately expand and compress spacetime in mutually perpendicular directions. Consequently, the distances in each tube fluctuate as the gravitational wave passes through them, causing the laser beam in each tube to arrive at the detector out of alignment. When they are recombined, the two beams no longer cancel each other out (the interference is partially constructive now), laser light is detected, and so, too, is the gravitational wave.

And this is what happened, for the first time in history, on the date of 14th September 2015. The aforementioned gravitational wave stretched and squeezed the 4-km long tubes in the Laser Interferometer Gravitational-Wave Observatory (LIGO) in the USA by one-thousandth the width of a proton! This was the first ever-direct detection of a gravitational wave, GW150914, yielding the 2017 Nobel Prize for Physics.

Although this first gravitational wave detection was monumental, it appears to have arrived alone, with no electromagnetic counterpart. There was one possible exception, however. The Fermi satellite reported measuring a weak gamma-ray burst from a region of sky that included GW150914. However, other space telescopes did not corroborate this. Stellar-mass black hole binary mergers are not expected to have large amounts of orbiting matter and so are not expected to produce gamma-ray bursts. Therefore, the Fermi signal is not regarded as a robust detection. It remains controversial, but we may yet discover an electromagnetic signal from merging black holes.

Since that first detection, one other gravitational wave detection was made with LIGO in 2015 and then four new ones arrived in 2017. All but one of these was similar to the first one, signifying two merging black holes. The one oddball, however, was incredible; GW170817 was the first binary neutron star to be detected. First the gravitational waves rippled past Earth, then 1.7 seconds later the light arrived. Light is expected when two neutron stars merge or a neutron star and a black hole merge. The Fermi satellite measured this light in the form of a gamma-ray burst, without controversy this time! My colleagues at University College London (UCL) Mullard Space Science Laboratory used UCL’s Ultraviolet and Optical Telescope on board the Swift satellite to catch GW170817’s ultraviolet light. All told, some 70 observatories measured the electromagnetic signals arriving from the neutron star merger that corresponded to the gravitational waves seen on 17th August 2017. It was the first cosmic event observed in both gravitational waves and light, representing a monumental step for multi-messenger astronomy.

The light from this event was classed as a kilonova, which is 1000 times brighter than a classical nova (an explosion on the surface of a white dwarf star). A supernova, meanwhile, is ten to one hundred times brighter than a kilonova. Kilonovae had previously been observed, but this landmark multi-messenger event demonstrated for the first time that kilonovae can be unambiguously caused by merging neutron stars.

10_031111Fig. 3 shows that gravitational waves have a frequency-dependent spectrum, similar to the electromagnetic spectrum. The terrestrial interferometers, such as LIGO, are sensitive to very short wave periods that correspond to the smallest objects. Therefore, gravitational astronomy is still in its infancy. Its next stage is to go to space, where the Laser Interferometer Space Antenna (LISA) will consist of three spacecraft, each a million miles apart, equivalent to a larger, longer-period version of LIGO’s 4-km tubes. While LIGO is sensitive to stellar mass black holes merging, LISA will be sensitive to supermassive black holes merging and capturing other objects. LIGO has to contend with confounding factors that cannot be removed, such as seismic noise and continental drift, whereas LISA will be free from this terrestrial interference when in space. LISA’s launch is planned for 2034, so we will have to wait for this new window on the Universe to open up for us.

Imprints of gravitational waves are expected in electromagnetic light. If they exist, primordial gravitational waves should be imprinted in the polarization of the Cosmic Microwave Background (CMB). Light consists of a coupled oscillating electric field (E-mode) and magnetic field (B-mode). B-modes in the CMB can be created by primordial gravitation waves arising from cosmic inflation: the exponential expansion of the Universe that came before and set up the Big Bang. It would be our first evidence that gravity is an inherently quantum force. In 2014, a team of scientists, using the Background Imaging of Cosmic Extragalactic Polarization (BICEP2) telescope at the South Pole, controversially reported that they saw such an imprint, but the claim was retracted a year later. BICEP3, along with other high-altitude telescopes, continue this search.

Gravitational waves will also perturb the distance between the Earth and pulsars in our Galaxy. Looking at multiple pulsars in different directions and timing any delays in their radio signals gives even longer-period gravitational wave sensitivity than LISA: searching for gravitational waves on a Galactic scale. Consequently, astronomers hope that pulsar-timing arrays will detect binary supermassive black holes in distant galaxies.


We need to “look” using more than just light to compile a comprehensive census of the Universe and to further our understanding of it. Gravitational waves are like sound waves, allowing us to “listen” to what the Universe is telling us. Detecting cosmic rays and neutrinos is like particles being detected in our nose, so we also get to “smell” what the Universe is like. These different ways of probing the Universe are like different senses, giving us different but complementary information about what makes up our cosmic reality.

Neutrinos were not observed with GW170817, probably because of its great distance and the fact that its jets were not pointing at Earth. If they had been, it would have been the ultimate multi-messenger astronomy signal, enabling us to detect all three types of message: gravitational waves, electromagnetic radiation and particles. Measuring all three from a neutron star merger would help us learn about the interiors of neutron stars and how dense matter can be before it collapses into a black hole.

2017 was an auspicious year for multi-messenger astronomy. The first event involving gravitational waves and light, GW170817, was observed five weeks before the first event involving neutrinos and light, IceCube-170922A. We are looking forward to the next message from the Universe, There is huge promise for 2019. LIGO is expected to start a new science run in February 2019. Nearly every telescope on Earth and in space has proposed follow-up observations of new anticipated gravitational detections. One of the big unanswered questions that could be answered next year is the rate of neutron star mergers – predictions range from 1 to 10. The origins of many of the elements in the periodic table still need to be pinned down. 2019 could shed further light (and particles and waves) on fundamental physics and even reveal new astrophysical phenomena.