University of Exeter
25th Anniversary
2001 – 2026
From protoplanetary discs to exoplanet atmospheres,
two and a half decades of pushing the frontiers of the Universe.
History
Milestones, discoveries, and the people who made them happen.
The University of Exeter Astrophysics Group is established within the Department of Physics & Astronomy, anchored by Professor Tim Naylor's stellar astrophysics programme. The group sets out to build world-class expertise in star formation, young stellar populations, and protoplanetary disc evolution.
Harries, Hilditch & Howarth use ten eclipsing binary stars — pairs of stars that periodically pass in front of each other — to measure a precise geometric distance to the Small Magellanic Cloud, one of the Milky Way's nearest galactic neighbours. Because eclipsing binaries reveal their own sizes and temperatures directly from how their brightness dips, they act as reliable "cosmic yardsticks" that become a benchmark rung on the ladder used to measure the scale of the Universe.
Bate, Bonnell & Bromm publish a seminal MNRAS paper presenting the first large-scale simulation to follow the collapse of a turbulent molecular cloud all the way to the formation of a realistic cluster of stars and brown dwarfs — directly predicting the initial mass function from first principles.
Bonnell, Bate & Vine simulate the collapse of a molecular cloud into a cluster of over 400 stars, revealing that clusters build up hierarchically — small groups of stars merging into ever-larger clusters — rather than forming in one smooth collapse. This crowded birth environment makes close stellar encounters common, and the team shows these can truncate discs, harden binaries, and even eject planets from forming systems entirely.
Bunker, Stanway, Ellis & McMahon mine the Hubble Ultra-Deep Field — then the deepest image of the Universe ever taken — to count galaxies as they appeared roughly a billion years after the Big Bang. They find the Universe was forming stars around six times more slowly at this early epoch, raising a puzzle over whether these faint galaxies alone could have supplied enough light to reionise the Universe.
Hatchell and collaborators complete an exhaustive submillimetre survey of the Perseus molecular cloud, one of the nearest active stellar nurseries, detecting 91 newly forming stars and cores. Combined with a map of the cloud's gas, the survey shows dense filaments are where new stars overwhelmingly form — helping establish the now-standard picture of filamentary star formation.
Kurosawa, Harries & Symington build detailed models of the gas falling onto young "T Tauri" stars and the winds blowing from their discs, tracing how both shape the star's telltale hydrogen emission line, Hα. Their models reproduce almost every profile seen in real young stars, clarifying when this single spectral fingerprint can be trusted as a measure of how fast a star is gathering mass.
Naylor & Jeffries develop a rigorous statistical method for fitting a star cluster's brightness and colour to stellar models, producing an age estimate with a genuine, statistically meaningful uncertainty rather than a rough by-eye fit. Released as free software, the technique became a widely used community tool for pinning down the ages of star-forming regions.
Building on their survey of the Perseus cloud, Hatchell and colleagues classify each newly discovered protostar by evolutionary stage, using its brightness across infrared and submillimetre wavelengths as a kind of developmental clock. From the numbers caught at each stage, they estimate that young stars spend under a million years in their earliest, most deeply embedded phase.
In the third instalment of their Perseus survey, Hatchell, Fuller & Richer hunt for the powerful jets of gas — molecular outflows — that newborn stars blast into their surroundings, finding them in most of the cloud's protostars. Outflow strength tracks a young star's mass and brightness, but can't alone explain how quickly protostars shed their gas envelopes, pointing to other processes at work.
Daniel Price releases splash, a free, interactive tool for visualising the output of smoothed particle hydrodynamics simulations — the particle-based technique used to model everything from star formation to colliding galaxies. Turning scattered simulation particles into a meaningful picture is surprisingly hard, and splash became, and remains, one of the standard tools of the trade across the simulation community.
Mayne & Naylor apply a consistent set of stellar models across eleven well-known star-forming regions to derive their distances and ages on a single, statistically robust footing, revising the accepted age ordering of several clusters. Their improved distance to the iconic Orion Nebula Cluster helped settle a long-running disagreement between competing measurements.
Bate publishes the first star cluster formation simulation to include full radiative transfer feedback, resolving long-standing questions about the role of accretion heating in setting the stellar initial mass function. The results appear in MNRAS and attract wide attention.
Pinte, Harries and an international team of code developers put seven independently written radiative transfer simulations — including Exeter's own TORUS code — through deliberately punishing test cases, modelling how light travels through the thick, dusty discs around young stars. Despite using completely different methods, the codes agree closely, giving the community confidence to trust them against real observations.
Brunt, Heyer & Mac Low compare radio observations of molecular clouds to turbulence simulations, using a statistical technique that pinpoints the physical scale at which the gas is being stirred up. Only simulations driven at scales comparable to, or larger than, the clouds themselves match what's observed, implicating galaxy-scale processes like supernovae rather than individual young stars.
Dobbs, Theis, Pringle & Bate simulate the famous "Whirlpool" galaxy M51 and its smaller companion, reproducing its sweeping spiral arms in remarkable detail — down to individual kinks seen in real images. The pattern turns out not to be a fixed, slowly rotating wave as classic theory predicted, but one constantly reshaped by the ongoing gravitational tug-of-war with its companion.
Bressert and a large international team, including Exeter's Jenny Hatchell, combine Spitzer surveys of nearby star-forming regions to map how tightly packed young stars are around the Sun. Rather than two distinct populations, they find a single smooth continuum of stellar crowding, with only a modest fraction of stars forming in dense environments where their discs and future planets might be disrupted.
Knigge, Baraffe & Patterson piece together how "cataclysmic variables" — close binaries in which a dense white dwarf siphons gas from a companion star — evolve over billions of years, using the donor stars' sizes as a fossil record. Their revised model resolves several long-standing puzzles and became a standard reference for understanding the endpoints of binary star evolution.
Sing and collaborators use the Hubble Space Telescope to measure how starlight filters through the atmosphere of hot Jupiter HD 189733b across the full optical and near-ultraviolet spectrum. The light scatters in a pattern consistent with a high-altitude haze blanketing the entire visible atmosphere — one of the first clear demonstrations that exoplanet atmospheres can be shrouded in haze.
Baruteau, Meru & Paardekooper simulate a proposed way of making giant planets — the direct gravitational collapse of clumps within an unstable disc — and track what happens next. The resulting planets migrate inward remarkably fast, meaning worlds born far from their star this way are unlikely to stay there, posing a challenge for explaining directly imaged planets like those around HR 8799.
Hennebelle & Chabrier derive an analytical theory predicting how quickly a turbulent molecular cloud converts its gas into stars. Unlike earlier recipes relying on an arbitrary density threshold, their theory lets the star formation rate rise smoothly with gas density and reproduces rates observed across a wide range of environments, becoming a widely used benchmark.
Matthew Bate publishes the largest radiation-hydrodynamical simulation of star cluster formation completed to date, following a molecular cloud's collapse into 183 individual stars and brown dwarfs. Fully accounting for the heat released as gas falls onto forming stars, the simulation reproduces the observed balance of stars to brown dwarfs far better than earlier models, showing gravity, hydrodynamics and radiative heating together shape the stars around us.
Aigrain, Pont & Zucker devise a simple method for predicting how much a star's own magnetic activity — starspots rotating in and out of view — will disturb the tiny velocity wobbles astronomers use to detect orbiting planets, using nothing more than the star's brightness variations. Now known as the "FF' method," it remains a standard tool for planet hunters working with active stars.
Vigan, Patience and an international team directly image 42 young, nearby massive stars in search of giant planetary companions at distances tens to hundreds of times the Earth-Sun separation. They estimate that roughly 6-19% of these stars host at least one giant planet in this outer zone, a benchmark figure for testing how such wide-orbit worlds form.
Dobbs & Pringle track individual giant molecular clouds — the vast gas reservoirs stars form from — through an entire galactic-disc simulation, from birth to dispersal. Rather than tidy, self-contained objects, the clouds have messy histories of merging and splitting, with typical lifetimes of a few tens of millions of years and a star formation efficiency of only about 1%.
Pont, Sing and colleagues stitch together years of Hubble and Spitzer observations into the most complete atmospheric spectrum yet assembled for hot Jupiter HD 189733b, dominated by scattering from a high-altitude haze rather than clear skies. Their proposal that hot Jupiters split broadly into "clear" and "dusty" types went on to shape how the whole field interprets exoplanet atmospheres.
Bell, Naylor, Mayne, Jeffries & Littlefair derive fresh ages for thirteen of astronomy's best-studied young star-forming regions using two independent methods that, for the first time, agree with each other. The surprising result — that many regions are up to twice as old as long assumed — implies planet-forming discs survive considerably longer than previously believed.
De Rosa, Patience and collaborators combine sharp adaptive-optics imaging with a search for co-moving companion stars to conduct the largest-ever survey of stellar companions around nearby A-type stars, resolving 137 companions including 64 newly discovered. Roughly two-thirds of these stars turn out to have at least one stellar companion — an essential benchmark for how massive stars form in pairs.
Mayne et al. publish the first paper applying the UK Met Office Unified Model to exoplanet atmospheric dynamics in A&A. The paper establishes a uniquely rigorous GCM framework for hot Jupiter circulation and day-night heat redistribution, spawning a decade of follow-up work.
Amundsen and colleagues stress-test the radiation physics at the heart of Exeter's effort to adapt the Met Office weather model for exoplanet atmospheres, comparing its fast approximations against far slower, exact calculations. Pinning down where these shortcuts introduce errors — sometimes by a factor of two — quietly underpins the accuracy of every hot-Jupiter simulation the group has produced since.
Dobbs & Baba write a major invited review pulling together decades of theory and observation on why spiral galaxies look the way they do. They argue that spiral arms, outside barred galaxies, are not fixed patterns but short-lived, recurring features that wind up over time — a comprehensive statement that became a standard reference for anyone studying galactic structure.
Matt, Brun, Baraffe, Bouvier & Chabrier derive a physically motivated formula for how a star's magnetised wind brakes its rotation over billions of years, and how that braking depends on mass. The model explains a long-standing puzzle in NASA Kepler data — why low-mass stars stay fast rotators far longer than Sun-like ones — and became a core ingredient of "gyrochronology," estimating age from spin.
Kataria and colleagues build three-dimensional climate simulations of WASP-43b, a hot Jupiter that orbits its star once every 19.5 hours, and test their predictions directly against Hubble spectroscopy. Without any fine-tuning, the model correctly predicts the timing and shape of the planet's dayside spectrum — a striking validation of the toolkit later used to interpret JWST observations.
Baraffe, Homeier, Allard & Chabrier publish the BHAC15 tracks in A&A: updated evolutionary models for low-mass stars and brown dwarfs that incorporate state-of-the-art atmospheric boundary conditions. Instantly adopted as the community standard for interpreting low-mass stellar photometry, they accumulate thousands of citations.
Sing, Fortney and a large international team assemble the most complete comparative study yet of ten hot Jupiter exoplanets, spanning visible light to infrared. Resolving a field-wide disagreement, they show planets span a continuum from clear to cloudy skies, and that clouds and hazes — not missing water — explain weak spectral signals. Published in Nature, it became one of the most cited works in exoplanet science.
Tremblin and colleagues propose a different explanation for why brown dwarfs' colours change abruptly as they cool: rather than patchy clouds breaking apart, a chemical instability similar to processes driving convection in Earth's oceans can naturally explain the same shift. The proposal offered a genuinely new physical picture for interpreting cooling brown dwarfs and directly imaged giant planets.
Drummond and colleagues tackle a subtle flaw in hot Jupiter atmosphere models: chemistry and temperature had normally been calculated separately, despite constantly influencing one another. Coupling the two together for the first time using Exeter's ATMO code, they find it can shift modelled temperatures by up to 100 degrees — enough to change how a real observed spectrum should be interpreted.
Brun & Browning, including Exeter's Matthew Browning, write a comprehensive review connecting what we know about the Sun's magnetism to the wider population of magnetic stars across the Galaxy. Pulling together observations, dynamo theory and cutting-edge simulations, the review became a standard reference for the "solar-stellar connection."
De Wit, Wakeford and an international team use Hubble to search for atmospheres on the four TRAPPIST-1 planets sitting in or near their star's habitable zone. They rule out, with high confidence, a puffy, hydrogen-dominated atmosphere for three of the four — strengthening the case for compact, rocky atmospheres and setting the stage for the JWST follow-up observations that continue today.
Exeter PhD student Jessica Spake leads an international team to the first-ever detection of helium in the atmosphere of exoplanet WASP-107b, a low-density super-Neptune 200 light-years away. The signal — observed in the infrared using HST — is so strong it implies an extended atmospheric envelope stretching tens of thousands of kilometres into space. Published in Nature, the technique paves the way for atmospheric characterisation of Earth-sized worlds with JWST.
Price, Wurster and a large team publicly release Phantom, a fast, modular code for simulating gas, dust and magnetic fields, developed over more than a decade. Built to handle everything from planet birth in swirling discs to gas falling into black holes, it has become one of the most widely used simulation codes in the field, powering much of Exeter's own work on protoplanetary discs.
Contreras Pẽna, Morrell, Naylor and collaborators catch a young star, Gaia 17bpi, in the act of a rare growth spurt — gas and dust from its surrounding disc crashing onto the star and bulking up its mass. It's only the 25th star of this type ever found, and the first time anyone has watched one of these events unfold in both visible and infrared light at once. The findings, published in The Astrophysical Journal, help explain how young stars put on weight as they grow — and what that might mean for the planets forming around them.
Wurster & Li review a tricky problem in star formation theory: models with realistic magnetic fields tend to brake away so much of a collapsing cloud's spin that the discs needed to form planets shouldn't exist at all — the "magnetic braking catastrophe." Their review of the proposed solutions, from misaligned fields to more realistic gas physics, became a key reference for modelling how discs come into being.
Professor Sasha Hinkley is selected to lead a large JWST Early Release Science (ERS) programme targeting the directly imaged planetary-mass companion VHS 1256 b. The programme will be among the very first JWST observations executed after commissioning, placing Exeter at the forefront of the JWST era.
Amard and colleagues compute an extensive public grid of stellar evolution models that, for the first time, self-consistently track how a star's rotation affects its structure — and how that rotation itself evolves — across a wide range of masses and compositions. Validated against real star cluster data, the grid is designed for use with Gaia, TESS and PLATO to determine stellar ages across the Galaxy.
Phillips, Tremblin, Baraffe and colleagues release ATMO 2020, an updated public set of models describing the atmospheres and long-term cooling of the coldest brown dwarfs and self-luminous giant exoplanets. Freely released to the community, it quickly became a standard reference grid used to interpret real observations, including many taken with JWST.
Kraus, Kreplin, Young, Bate and a large international team use powerful interferometric imaging to capture direct evidence of "disc tearing": a young triple-star system, GW Orionis, whose gravity has ripped its planet-forming disc into separate, misaligned rings that cast shadows across the warped disc between them. Long predicted but never clearly observed before, it offers a natural explanation for planets found on strange, tilted orbits. Published in Science.
As Jupiter and Saturn draw closer together in the sky than at any point since Galileo first turned a telescope skyward — a "Great Conjunction" not seen this close since 1623, and one that won't recur until 2080 — the Exeter Astrophysics Group and Exeter Science Centre brings the spectacle to a global audience. A live-stream from a telescope on the Physics Building roof draws thousands of viewers from around the world, while an accompanying series of explainer videos on how to observe the event and the science and history behind it rack up hundreds of thousands of views, reaching curious adults and children across the globe.
The James Webb Space Telescope launches on 25 December 2021. Exeter astrophysicists have secured Guaranteed Time and Early Release Science programmes to exploit it for exoplanet atmospheres, protoplanetary disc science, and stellar interiors — positioning the group at the leading edge of the new era.
Miles et al. (led by Exeter's ERS team, PI Hinkley) publish the first JWST NIRSpec spectrum of a directly imaged planetary-mass object — VHS 1256 b — in ApJL. The spectrum reveals unprecedented chemical detail: water, methane, CO, CO₂, and silicate clouds, transforming our understanding of giant exoplanet atmospheres. NASA, ESA and the RAS issue major press releases; the paper attracts global media coverage.
As part of the Transiting Exoplanet ERS programme, Exeter researchers contribute to a suite of Nature papers presenting the most detailed chemical characterisation of a transiting hot Saturn yet achieved — detecting CO₂ for the first time in an exoplanet atmosphere.
Carter, Hinkley and the JWST Early Release Science team capture the first images of an exoplanet ever taken by JWST, imaging giant planet HIP 65426 b across seven filters from 2 to 16 microns — including the first-ever direct detection of an exoplanet beyond 5 microns. The observations show JWST outperforming pre-launch predictions by up to a factor of ten, demonstrating its exceptional promise for imaging planets around other stars.
The Exeter Astrophysics Group celebrates its 25th anniversary. From its founding in 2001 as a small stellar-astrophysics cluster, the group has grown into an internationally recognised centre spanning star formation, interferometry, exoplanet direct imaging, atmospheric modelling, and stellar evolution — with researchers contributing to JWST, ALMA, VLTI, CHARA, and preparations for the Terra Hunting experiment and LSST.
Research
A selection of the papers that have defined our research.
The first large-scale SPH simulation to follow the collapse of a turbulent molecular cloud into a star cluster, directly reproducing the observed stellar initial mass function — including brown dwarfs — from first principles.
View paperThe first radiation-hydrodynamic star cluster formation simulation, demonstrating that radiative heating from accreting protostars suppresses disc fragmentation and sets the characteristic stellar mass — a breakthrough in understanding the origin of the IMF.
View paperHST observations revealing the featureless, grey transmission spectrum of HD 189733b, interpreted as evidence for high-altitude hazes and scattering aerosols — a pivotal result that shifted the field's understanding of aerosol physics in exoplanet atmospheres.
View paperAperture-masking interferometry at Keck reveals point-source emission interior to the dust-depleted cavity of LkCa 15's transitional disc — the first direct interferometric evidence for a planet still assembling within a disc.
View paperThe foundational paper adapting the UK Met Office Unified Model for exoplanet atmospheric dynamics. It establishes the Exeter GCM framework for hot Jupiter circulation, demonstrates robust jet-stream formation, and opens a new programme of exoplanet climate modelling.
View paperUpdated stellar and substellar evolutionary models incorporating state-of-the-art atmosphere grids. The BHAC15 tracks are now the community standard reference for interpreting photometry and spectra of low-mass stars, brown dwarfs, and giant planets.
View paperA comparative Hubble and Spitzer survey of ten hot Jupiters spanning the full optical-to-infrared range, resolving a field-wide debate over why some exoplanets show unexpectedly weak water absorption. The results reveal a continuum from clear to cloudy atmospheres, showing that clouds and hazes — not missing water — explain the weakest spectral signals, and becoming one of the most cited papers in exoplanet atmospheric science.
View paperMonte Carlo radiative transfer modelling of the thermal structure of protostellar envelopes, demonstrating how carbon-chain molecules can survive in warm gas close to the central protostar — with implications for disc chemistry and the origin of complex organics.
View paperThe first detection of helium in an exoplanet atmosphere, found via a strong absorption signal in the atmosphere of the low-density super-Neptune WASP-107b. Led by then-Exeter PhD student Jessica Spake, the result established helium as a powerful new tracer of atmospheric escape, now used to study atmospheric erosion across the exoplanet population.
View paperLarge-scale SPH simulations examining how the concurrent processes of pebble accretion, gas accretion, and orbital migration shape the architecture of emerging planetary systems — with direct comparison to the observed period-mass distribution of exoplanets.
View paperALMA high-resolution observations reveal a circumplanetary disc surrounding PDS 70c — the first unambiguous detection of a disc feeding material onto a young, still-forming giant planet. Exeter's Stefan Kraus contributes to the interferometric analysis confirming the disc's structure and mass.
View paperThe first JWST spectrum of a directly imaged planetary-mass companion, revealing water, methane, CO, CO₂, and silicate cloud features in unprecedented detail. Led by Exeter's JWST ERS team under PI Sasha Hinkley, this was one of the highest-profile early JWST results.
View paperNIRSpec observations of the transiting hot Saturn WASP-39b deliver the first unambiguous detection of CO₂ in an exoplanet atmosphere — a landmark result establishing JWST's power for atmospheric chemistry and demonstrating the feasibility of carbon-cycle studies on other worlds.
View paperBroader Impact
The methods we develop to study the Universe find unexpected applications here on Earth.
To simulate the atmospheres of distant exoplanets, Exeter researchers partnered with the Met Office to adapt their flagship Unified Model — the same code used to produce the UK's daily weather forecast. The collaboration runs both ways: insights from the extreme conditions of hot Jupiters are feeding back into how we model circulation and cloud formation on our own planet, sharpening the tools used to study the Earth's changing climate.
The radiative transfer codes Exeter developed to model light scattering in protoplanetary discs are now being turned on human tissue. Working with clinicians at the Royal Devon & Exeter Hospital, the team has adapted these codes to detect calcium deposits in breast tissue — a key early indicator of cancer, where early detection raises five-year survival rates from 15% to 90%. The same modelling is being used to simulate photothermal therapy for skin cancer, predicting how gold nanoparticles heat and destroy tumour cells under near-infrared light, with human clinical trials in progress.
Find out moreHigh-resolution humidity measurements are the missing ingredient in accurate regional weather forecasting. Brunt is applying radio interferometry techniques — honed on molecular clouds and star-forming regions — to a new problem: measuring the bending of ADS-B signals routinely broadcast by commercial aircraft for air traffic control. Variations in how these signals refract reveal detailed water vapour profiles through the atmosphere. With thousands of flights crossing UK airspace daily, this approach could deliver millions of humidity measurements per day at modest instrument cost.
Find out moreMonte Carlo radiative transfer codes originally built to trace starlight through interstellar dust have been put to work modelling artificial light at night — simulating in three dimensions how streetlight emission spreads, reflects, and accumulates across a city. Working alongside world-leading ecologists at Exeter's Penryn Campus, and harnessing cutting-edge nighttime remote sensing imagery, the team links these physical light models to how nocturnal animals actually behave, mapping effects on plant photosynthesis, melatonin cycles, and wildlife activity at the fine spatial resolution ecologists need. The project's findings are now feeding directly into policy, working with local councils to help shape lighting strategies that keep streets safe while giving wildlife the dark it needs.
Find out moreLooking Ahead
The golden age of astrophysics is not behind us — it is just beginning. With the JWST delivering its first transformative data, the Extremely Large Telescope on the horizon, and a new generation of researchers joining the group, the next quarter-century promises to build on 25 remarkable years of discovery and take us even further.
JWST has opened a new observational window onto the chemistry and climate of worlds beyond our Solar System. Exeter researchers are leading programmes to characterise the diversity of planetary atmospheres — from hot Jupiters to potentially habitable rocky planets — with a level of precision that was unimaginable a decade ago.
The origin of planetary systems remains one of the deepest unsolved problems in astronomy. Using ALMA, VLTI/GRAVITY, and next-generation interferometers, our group continues to image the birth of planets at resolutions that resolve individual au-scale structures within protoplanetary discs.
Exeter is helping build the instruments that will define astronomy's next chapter. Professor Stefan Kraus leads development of MIRC-X and BIFROST — next-generation interferometric imagers for the CHARA Array and ESO's VLTI, designed to directly image exoplanets and young stars at unprecedented resolution. Alongside continued JWST observing time, the group is a founding partner in the Terra Hunting Experiment, a decade-long survey using the HARPS3 spectrograph to search for true Earth-mass, Earth-temperature twins.
"We stand on the shoulders of twenty-five years of brilliant science, and we look upward."