skip to primary navigationskip to content


Earliest galaxies in the Universe’s history spun like the Milky Way


Artist impression of rotation in a galaxy in the early Universe, credit: Institute of Astronomy, Amanda Smith.

Astronomers have looked back to a time soon after the Big Bang, and have discovered swirling gas in some of the earliest galaxies to have formed in the Universe. These ‘newborns’ – observed as they appeared nearly 13 billion years ago – spun like a whirlpool, similar to our own Milky Way.

An international team led by Dr Renske Smit from the Kavli Institute of Cosmology at the University of Cambridge used the Atacama Large Millimeter/submillimeter Array (ALMA) in Chile to open a new window onto the distant Universe, and have for the first time been able to identify normal star-forming galaxies at a very early stage in cosmic history with this telescope. The results are reported in the journal Nature, and will be presented at the 231st meeting of the American Astronomical Society.

Light from distant objects takes time to reach Earth, so observing objects that are billions of light years away enables us to look back in time and directly observe the formation of the earliest galaxies. The Universe at that time, however, was filled with an obscuring ‘haze’ of neutral hydrogen gas, which makes it difficult to see the formation of the very first galaxies with optical telescopes.

Smit and her colleagues used ALMA to observe two small newborn galaxies, as they existed just 800 million years after the Big Bang. By analysing the spectral ‘fingerprint’ of the far-infrared light collected by ALMA, they were able to establish the distance to the galaxies and, for the first time, see the internal motion of the gas that fuelled their growth.


Two galaxies 800 million years after the Big Bang (6% of the current age of the Universe) as seen by ALMA (inset panels). The color gradient in the sources shows the motion of the gas detected by ALMA, indicative of rotation. credit: Hubble (NASA/ESA), ALMA (ESO/NAOJ/NRAO), P. Oesch (University of Geneva) and R. Smit (University of Cambridge).

“Until ALMA, we’ve never been able to see the formation of galaxies in such detail, and we’ve never been able to measure the movement of gas in galaxies so early in the Universe’s history,” said co-author Dr Stefano Carniani, from Cambridge’s Cavendish Laboratory and Kavli Institute of Cosmology.

The researchers found that the gas in these newborn galaxies swirled and rotated in a whirlpool motion, similar to our own galaxy and other, more mature galaxies much later in the Universe’s history. Despite their relatively small size – about five times smaller than the Milky Way – these galaxies were forming stars at a higher rate than other young galaxies, but the researchers were surprised to discover that the galaxies were not as chaotic as expected.

“In the early Universe, gravity caused gas to flow rapidly into the galaxies, stirring them up and forming lots of new stars – violent supernova explosions from these stars also made the gas turbulent,” said Smit, who is a Rubicon Fellow at Cambridge, sponsored by the Netherlands Organisation for Scientific Research. “We expected that young galaxies would be dynamically ‘messy’, due to the havoc caused by exploding young stars, but these mini-galaxies show the ability to retain order and appear well regulated. Despite their small size, they are already rapidly growing to become one of the ‘adult’ galaxies like we live in today.”

The data from this project on small galaxies paves the way for larger studies of galaxies during the first billion years of cosmic time. The research was funded in part by the European Research Council and the UK Science and Technology Facilities Council (STFC). 

Galaxy visualisation, credit: R. Crain (LJMU) and J. Geach (U.Herts)

Other versions of the video may be found at


Renske Smit et al. ‘Rotation in [C II] - emitting gas in two galaxies at a redshift of 6.8’. Nature (2018). DOI: 10.1038/nature24631


Renske Smit:


Using GAIA to detect low frequency gravitational waves

The GAIA satellite

The GAIA satellite

A group of Cambridge astronomers, including Anthony Lasenby from the Cavendish Astrophysics Group and Kavli Institute for Cosmology, have made the first investigation of the sensitivity of the GAIA satellite to ultra-low frequency gravitational waves produced by supermassive black hole binaries. (See Moore, Mihaylov, Lasenby & Gilmore, Phys. Rev. Lett. 119, 261102, 2017.)

The method is similar to that used in Pulsar Timing Arrays, except that instead of the gravitational wave modifying the apparent frequency of a pulsar, it modifies the apparent positions of stars observed by GAIA, making them oscillate with a characteristic pattern on the sky, An example of this is shown in the figure,



in which a gravitational wave is travelling from one pole of the celestial sphere, and the black and red lines indicate the induced apparent motions of the stars within a hemisphere (exaggerated by a large factor to make them visible), corresponding to the two possible polarisations of the wave. Using a method which allows the information in the star motions to be compressed by a factor of 10^6, the study shows that it will be possible to search for individually resolvable waves within the full GAIA data set of many billions of stars, and that the sensitivity of GAIA could be comparable to that of current or near-future pulsar timing arrays over a slightly wider frequency band.

The paper was selected as the cover article for a recent issue of Physical Review Letters, and is accompanied by an Editor's Choice Focus article.



Rosie Bolton

Please, welcome to the stage... Dr Rosie Bolton!
Rosie Bolton
Full-size image: 362 KB | View imageView Download imageDownload

''Our experiment is the Universe'' Dr Rosie Bolton said

The International Conference for High Performance Computing, Networking, Storage and Analysis, held in Denver (Colorado) in November 2017, was the perfect subliminal stage to highlight the potential of the the Square Kilometre Array (SKA) project, thanks to the wise projection of visually stunning pictures of stars and telescopes.


Please, welcome to the stage... Dr Rosie Bolton!

One of the keynoters was Rosie Bolton, SKA Regional Centre Project Scientist and Project Scientist for the international engineering consortium designing the high performance computers (SKA SDP).

Our First Close Look at another Solar System

This blog entry has been written by Ellie Kershenbaum, a student from Sawston village college. 


In February 2017, the University of Liège and the University of Cambridge announced that about 39 light years away there is a system of seven Earth-sized planets orbiting an ultra-cool dwarf star called TRAPPIST-1. This discovery made news all over the world and for good reason. All planets in this system are temperate, meaning they are under certain geologic and atmospheric conditions to potentially host liquid water, although three are thought to receive the right temperature from the star to sustain water in liquid form. This suggests that there is a possibility that extra-terrestrial life exists there! The host star is only about 8% the mass of the Sun and is not much bigger than Jupiter.


All the planets in this system are called exoplanets. These are planets outside of our Solar System that orbit a star. The study of exoplanets is a very fast-moving and relatively new field. The first confirmed detection of an exoplanet was in 1992, while the first detection of an exoplanet orbiting a main sequence star (a Sun-like star) did not occur until 1995, and was made by Michel Mayor and Didier Queloz. There are many reasons why we should find and observe exoplanets. One reason is that we have to observe exoplanets in order to look for signs of extra-terrestrial life. Dr Sylvestre Lacour, a visiting astronomer at the Battcock Centre for Experimental Astrophysics, works on a small satellite called PicSat (1) that will detect exoplanets from space. He says that observing exoplanets can give us a better understanding of how planets form, by looking at different exoplanets and identifying different stages of a planet’s life. He also suggests that observing exoplanets could predict what could happen to the Earth in the future, for instance, the effects of climate change.


TRAPPIST-1 was discovered using transit photometry, which involves detecting a small regular loss of light emitted by a star, which shows that a planet has crossed the star as it moves along its orbit, blocking a small fraction of the light from reaching us. This is known as a transit. The larger the dip in light is, the more of the surface of the star the planet has covered, and therefore the bigger the planet is. This method has proven useful in detecting exoplanets, however there are some problems with it. For example, background noise, and changes in the activity of the star can slightly change the amount of radiation that is detected. This means that it is more difficult to detect the small, regular dips of light from the planet transiting.


Fig. 1: The transit light curves of all seven planets in TRAPPIST-1. (2)

There are other methods of detecting exoplanets. For instance, the gravitational pull of the planet orbiting the star causes the star to wobble. The radial-velocity method detects the blue shift (when the star moves closer to us) and the red shift (when the star moves away from us) of a star as it wobbles. One problem with this method is that the red/blue shift could actually be caused by a star that is further away, rather than a planet. Another method for detecting exoplanets is direct imaging. This consists of taking images of a planetary system to see if there are planets orbiting the star. The problem with this method is that the planet could be difficult to see, as the parent star’s light could overpower it. Therefore, it is only viable for bright planets in distant orbits around nearby faint stars. I spoke to Dr Samantha Thompson, a research associate from the Battcock Centre, who mainly works on the development of new instruments and techniques to detect and characterise exoplanets. She believes that developing high-contrast imaging, for the direct imaging method of detecting and characterising exoplanets, is extremely important and useful. This is because direct imaging is a potent way of characterising exoplanets. Rather than inferring characteristics (e.g. size, distance from star), as transit photometry and the radial velocity method do, direct imaging allows us to see them directly. High-contrast imaging, such as coronagraphy (viewing objects near a star by blocking its bright surface), would suppress starlight and preserve planet light. This would enable us to see the planet and its characteristics more clearly.

The inner three planets, TRAPPIST-1b, TRAPPIST-1c, and TRAPPIST-1d are thought to be too close to the star, so the heat energy from the star would evaporate any water. Similarly, the outermost planet, TRAPPIST-1h, is thought to be too far from the star, so the water would solidify, however, internal heating, caused by tidal friction between the planet and star, may increase the chances of liquid water. Three of the seven planets, TRAPPIST-1e, TRAPPIST-1f, and TRAPPIST-1g, are described as being in the Habitable Zone.  The Habitable Zone, or Goldilocks Zone, is the distance from the star at which water can exist in liquid form as it is not too close or too far from the star.


Fig. 2: The planets the TRAPPIST-1 system, three being in the habitable zone. (3)

Density measurements suggest that at least six of the seven planets are probably rocky in composition, which gives a better chance of life existing there. The low density of TRAPPIST-1f suggests it might have a volatile-rich composition (chemical compounds and elements with low boiling points) which could be in the form of an ice layer and/or an atmosphere. The planets in the system are all Earth-sized (except TRAPPIST-1d and TRAPPIST-1h which are between the sizes of Mars and Earth) and have a similar mass to Earth. The mass of a planet is not as simple to calculate as some other features such as its diameter when using transit photometry. However, there is a way of estimating the mass. Dr Amaury Triaud, a Kavli Exoplanet fellow at the Institute of Astronomy, is a co-author on the key paper recently published about the TRAPPIST-1 system (2). He explains that the mass of an exoplanet can be estimated by looking at the effect it has on other planets. The gravitational force of a planet can accelerate or decelerate another planet’s orbit which will make it transit slightly late or slightly early. This can be used to estimate a planet’s mass.


It may seem strange that a system with an ultra-cool dwarf star (a star that’s very different from our own) would be host to Earth-like planets, with chances of life, but it’s not. It’s been suggested that instead of trying to look for a system identical to ours we should start to look for different systems such as dwarf planetary systems. This is because the stars that are like our Sun represent only 15% of the stars in the Milky Way, more than half of those being in binary systems (a system in which two stars orbit around a common centre of mass, gravitationally bound to each other). Our star is therefore relatively unusual, and does not represent the other stars in the Milky Way. Another reason we should be looking at dwarf planetary systems is that it may be more likely for a rocky planet to orbit a low-mass star. Furthermore, transits are more prominent with a smaller star as the planet will block a larger fraction of light, meaning it is easier to detect. With a smaller star, transit periods are shorter as the orbits are closer, which also means it is more likely that we will be able to detect a transit. The more transits we can detect of the same planet, the more confident we can be about the properties of that planet.


Now for the big question, is there life on these planets? As well as having many orbiting planets, TRAPPIST-1 probably got a lot of news coverage because of the possibility of life existing there.  This is a very exciting topic that makes us think of all the science fiction films, TV shows, and books. It’s important for us to try to expand our knowledge about life in the universe as it could give us a clue to how life began on Earth.  There are a lot of factors that play a big role when it comes to the possibility of life. For example, the composition of the atmosphere. Detecting certain gases and chemical compounds in the atmosphere, such as carbon monoxide, water, or oxygen, would tell us how habitable the planet is, and how likely it is for life to evolve there. The detection of biologically produced gases in the atmosphere, such as methane, could be indicative of life. However, some indicators can be misleading. For instance, repeated stellar flares could result in oxygen in the atmosphere. The likelihood of life also depends on whether the planet is rocky, which can be determined by calculating its mass and volume, looking to see if the planet is relatively dense. However, it is possible that the planets (at least the ones closer to the star) are tidally locked, so that they don’t rotate, meaning that each point on the surface has either constant day or constant night. This could lead to one side of the planet being too hot to support life, and one too cold, which means life could only evolve at the rim that connects the dark and light side.


Fig. 3: A planet tidally locked with its star, one side facing the star constantly as it moves around its orbit. The planet does not rotate in its orbit around the star due to the star’s gravitational pull.

This discovery is very important for astrophysics and astrobiology, but there are still things that we don’t know or can’t be sure about, for instance the chemical composition and atmospheric structures of the planets. For this reason, the James Webb Space Telescope, which will launch in 2018, and the Hubble telescope, will collect further observations on the TRAPPIST-1 system from space. There are also plans of finding more Earth-like exoplanets in dwarf planetary systems, including ultra-cool dwarf stars.


So, why should we care about this system? The TRAPPIST-1 system could improve our understanding of star and planet formations and is an important target for the search for extra-terrestrial life, to further our understanding of the conditions that life requires. Dr Triaud suggests that the discovery of TRAPPIST-1 could indicate how common life is in the universe. We might be unique and rare, or surrounded by thriving alien life we just haven't detected yet. There has been increasing evidence to suggest that Earth-like planets are not that uncommon. NASA’s Kepler space telescope has already released a recent mission catalogue which includes 219 new planet candidates, 10 of which being near-Earth size and orbiting in their star's habitable zone (4). If Earth-like planets in the habitable zone are not extremely rare, maybe life isn’t either. The TRAPPIST-1 system could also lead to other discoveries, constantly improving the instrumentation we use and the knowledge we have of our universe. So we probably should care.







On the scientific importance of the detection of gravitational waves

On the scientific importance of the detection of gravitational waves

Anthony Lasenby

Gravitational Waves were first predicted by Einstein in 1916, as an outcome of his General Theory of Relativity. They are waves in the metric of GR that propagate at the speed of light. This sounds straightforward, but in fact acceptance of these waves as being a prediction concerning real physical phenomena did not happen till much later. A major problem was being able to separate out coordinate effects from real physical effects, and also whether the singularities which seemed to occur after the passage of a wave in exact treatments, rather than the usual linearised treatment, were real, or were themselves coordinate artefacts.

By the early 1960s, however, gravitational waves were accepted as real phenomena, and attention moved to their detection. Early claims of detection using resonant bar detectors in the 1960s by Joe Weber were at a level of 'gravitational strain' - the dimensionless units in which the amplitude of a wave can be measured - that were much too large compared to what was thought possible from astrophysical sources. It had to wait to September of last year for the level of sensitivity of an instrument to reach that necessary to detect astrophysical phenomena. The experiment concerned is the Advanced LIGO instrument in the US, and a strong candidate event was found in the very first scientific run of the instrument. The event is coincident in time between the two detectors of LIGO, one in Washington State and the other in Louisiana, and matches extremely well the expected signal from the rapid inspiralling and 'ringdown' of two black holes, each with a mass near 30 Solar masses.

This measurement represents several 'firsts'. It is clearly the first detection of gravitational waves, but also the first detection of black holes, as well of course of black hole binaries, in this mass range. The final product of the black hole coalescence, which takes only a fraction of a second to complete, is a rotating black hole, whose spin can be measured directly from details of the gravitational waveform as the combined object relaxes towards the Kerr form of the metric. The measured spin is 0.7 of the maximum possible for a rotating black hole, and represents another first, since although we have had indirect measurements of black hole spin for many years, this is the first direct measurement.

Several other aspects of strong field relativity can be tested from the waveform, and also new limits set on any possible mass of the graviton, which if it did have mass would cause distortion of the signal as it propagated towards us. The astrophysical implications of the event witnessed, which of course requires the creation of a high mass black hole binary system at some point in the past, are also significant, and a crucial question will be the rate at which such events occur. Thus the possibility of finding equivalent events in future data, as well as the appearance of the other types of interactions (such as the coalescence of neutron star binaries, which should also be detectable), is keenly anticipated.

International Year of Light and Light-based Technologies


International Year of Light and Light-based Technologies

Aglaé Kellerer

The International Year of Light and Light-based Technologies involves artists and scientists across a large range of domains. It certainly also involves astronomers and to honour this year, we might temporarily rename our group the Laboratory of Light Pathology. Astronomers analyze light in order to understand the formation, evolution and subsequent fate of celestial bodies. While though this sounds obvious today, it seemed inconceivable in the past. 180 years ago, the French philosopher Auguste Comte argued that we would never be able to determine the chemical or mineralogical structure of planets other than Earth since we couldn't obtain geological samples (Auguste Comte, “The Positive Philosophy”). Nowadays our improved understanding of light-matter interactions allows precise analyses of planetary, stellar and inter-stellar compositions. 

One of the most extraordinary aspects of our research is the process of astronomical imaging itself. Consider a photon emitted by M33, the Triangulum Galaxy. If this is your favorite galaxy, you are quite right, because it certainly looks stunning (see Fig.1). Consider a photon emitted by a star in this galaxy 3 million years before it reaches your telescope. Moments before it is recorded by your detector, it is located on an immense sphere centered around M33, a sphere with a radius of 3 million light years.

Fig.1: M33, the Triangulum Galaxy. Image Credit & CopyrightRobert GendlerSubaru Telescope (NAOJ). Image data: Subaru Telescope, Robert Gendler, Brigham Young University Obs.Johannes Schedler

Now imagine a group of astronomers on Htrae, a small imaginary planet that orbits a system of two stars, on the other side of M33, 6 million light years from Earth. M33 has been a longtime favorite of Notwen, the president of Htrae's Physical Society. At this moment, a new instrument is being tested, an overwhelmingly large visor (OWL) - it has a diameter of 0.1m, very large indeed for such a small planet. Notwen has decided to aim it towards M33. From that point, the race is on. Who will collect the photon, you or Notwen?

Both your telescope and Notwen´s OWL have an - admittedly tiny - chance to record the photon. Both are 3 million light years away from the source of the photon, with your 0.2m diameter instrument the chance is 4 times larger. But if either your detector clicks or OWL does, the other detector´s chance to observe the photon vanishes immediately. Notwen is six million lights years away, but your detection of the photon has an instantaneous effect on its instrument, it wipes out its chance to register the photon.

How can we understand this seeming disregard of photons for space and time?

Newton was famously bothered by Gravity's action at a distance: "That gravity should be innate, inherent and essential to matter so that one body may act upon another at a distance through a vacuum without the mediation of any thing else by and through which their action or force may be conveyed from one to another is to me so great an absurdity that I believe no man who has in philosophical matters any competent faculty of thinking can ever fall into it." ( However, he also explained that "it is enough, that gravity does really exist, and act according to laws which we have explained, and abundantly serves to account for all the motions of the celestial bodies, and of our sea." (

Today we are in a similar situation with elementary particles. Quantum electrodynamics is both bewildering and successful. We use it to understand astronomical observations and to improve the performance of telescopes, and we wonder about its profound mysteries. Photons do not experience space and time as we do on our scales. Do the concepts of space and time merely emerge from large networks of particles?

Telescopes allow us to explore the frontiers of the Universe, to investigate what it is made of and how it evolved. Since the telescope was first invented, our knowledge of the laws of optics have been revolutionized, notably with the advent of Quantum Electrodynamics. Can we use novel quantum optical processes to improve the performance of our telescopes and, thereby, improve our knowledge of the Universe?


Fig.2: Notwen, the president of Htrae's Physical Society, tests the new overwhelmingly large visor (OWL) on M33. You observe the same target from your backyard.

Fig. 3: Notwen is six million lights years away, but your detection of the photon has an instantaneous effect on him: it wipes out his chance to register the photon.


Gottesman et al. ( and Riaud ( have independently proposed to use photon entanglement to increase the distance between the telescopes of an interferometer. Cotler and Wilczek ( just recently suggested to substantially widen the possibilities of interferometry by using entanglement. I suggested a scheme to overcome the diffraction limit at the expense of sensitivity by use of quantum non-destructive measurements (QND) (

QND detections are a radical breakthrough of modern optics: When photons arrive on a detector - e.g. the retina or a CCD chip - they interact with atoms, the energy of the photon is transmitted to the atom and the photon is thereby destroyed. In a non-destructive measurement, the photon is detected, but not destroyed. This is an active field of research, notably in the context of quantum computers, where photons are to be used as information carriers (flying qu-bits) between quantum gates ( and

As astronomers, we have every reason to celebrate the International Year of Light. Most of our knowledge about the Universe comes from light. New technologies help us extract the maximum information out of every collected photon. This will allow us to make discoveries that we do not yet imagine.

ALMA detects [CII] emission from a primeval galaxy at z~7

ALMA detects [CII] emission from a primeval galaxy at z~7

Rebecca Williams


Roberto Maiolino is leading a study using ALMA to detect the [CII] 158μm line and continuum emission in galaxies well within the epoch of re-ionisation at z~7. These galaxies have very low star formation rates (SFRs ~10 M¤/yr) and so are more representative of the galaxy population at high redshifts than those studied in previous investigations. The [CII] 158μm line is the dominant coolant line of the ISM and so can be a powerful tool for studying the ISM in high-z galaxies. Also at z > 1, the line is redshifted into the sub-mm atmospheric window allowing it to be detected using ALMA.


By targeting the [CII] line with ALMA we have detected clumps of neutral gas in the vicinity of a galaxy at z = 7.107 which are spatially offset from the primary galaxy previously identified through Lyα + UV continuum emission (which is tracing the ionised gas). This is illustrated in Fig1[1]. The line profile (shown in Fig2) shows that the [CII] emission is consistent with the redshift obtained from the Lyα emission (which appears artificially redshifted due to absorption of the blue shoulder from the intervening IGM indicated by the observed asymmetric line profile). The [CII] emission profile also appears much narrower than that of the Lyα, which has been previously observed [2] in an investigation of two Lyα emitters in the vicinity of BRI1202 (a QSO-SMG system). However, these observations represent ‘peculiar’ objects in an overdense environment that also hosts a quasar and hence are not representative of the bulk of the galaxy population. The recent ALMA results however are considered much more unique in the sense that it is the first time such a feature has been seen in a ‘normal’ galaxy.


Fig1. Image shows ALMA [CII] detection, which is spatially offset from the location of the primary galaxy traced by Ly alpha+UV continuum and indicated by the white cross.

Fig2.  Top: Extracted ALMA spectrum showing the [CII] emission line. Bottom: Lyα spectrum showing asymmetric profile. Note that the [CII] redshift is consistent with that previously confirm through Lyα emission. 

The spatial offset and spectral line differences between the [CII] emission (tracing neutral gas) and the Lyα emission (tracing ionised gas) are predicted by recent models of primeval galaxy formation. According to these models, molecular clouds in primeval galaxies are disrupted by stellar feedback whereas accreting/satellite clumps of neutral gas survive in the vicinity of the galaxy and are detectable through their [CII] emission. Fig3 shows the simulation of a primeval galaxy with a similar SFR to the observed galaxy (see Fig3 caption for details). It shows that little [CII] emission is expected at the position of the primary galaxy (shown as the colour image in Fig3) hence the majority of this [CII] emission is not expected to be as a consequence of in-situ star formation. Instead offset neutral clumps of gas emit [CII] (shown as contours in Fig3) due to irradiation from the primary ionised galaxy, which is consistent with our observation (see Fig1).



Fig 3. Simulation of a primeval galaxy at z~7.1. Colour shows the ionised gas (e.g. Lyα) tracing the primary galaxy, the black points show the UV continuum and the contours indicate the location of the neutral (i.e. [CII]) gas.


These results imply that the observations are probing galaxy formation at very early stages, which in turn will help in further constraining models of galaxy formation and identifying the key components in galaxy evolutionary scenarios.


For full details see: [1] Maiolino et al., 2015, MNRAS, submitted;  [2] Williams et al. 2014, MNRAS , 439, 2096