Science1 publisher2 min readPublished
MEGATRON simulations couple starlight and chemistry to link JWST galaxies with the Milky Way's oldest stars
University of Bath-led MEGATRON team published four papers arguing early-universe models must track starlight, gas and new elements together. Their simulations suggest simpler models understate how stellar radiation and chemistry shape the gas around young galaxies.
The Scientist · Science desk
What happened
- The simulations start from pristine gas with no heavy elements and follow the first stars, their radiation, their supernovae and the spread of new elements into later stars.
- At high resolution the team resolved structures in the surrounding gas that simpler models do not capture.
- The collaboration has been awarded 40 million processor hours on UK national supercomputers for its next generation of runs.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability A single simulation that produces both a young galaxy and the chemistry of its later stars can be checked against JWST and stellar surveys at once, a stricter test than either data set alone.
- constraint With one simulated galaxy, the papers cannot yet say how typical its gas structures are, so the conclusion describes one history until more are run.
- cost Physics at this fidelity needs a lot of national supercomputer time: about 4,570 processor-years for the next round alone.
Early-universe astronomy keeps two records that have been treated as separate views [3]. The James Webb Space Telescope sees young galaxies directly, while ancient stars in and around the Milky Way hold a chemical fossil record from which astronomers reconstruct the first stars [9]. Martin Rey of the University of Bath's physics department, a lead contributor to the collaboration [14], said, "The James Webb Space Telescope gives us a direct glimpse of the infant cosmos, while stellar archaeology allows us to study the relics of those earliest times in our own galactic neighborhood. MEGATRON provides a physical bridge between the two." [10]
One simulation does both jobs here. The code tracks gas motion, the spread of starlight and changing chemical concentrations together [5], so the stars that light up the young galaxy are the same stars whose supernovae scatter the elements later generations inherit [8]. I like the design, because one simulated history has to answer to both kinds of observation at once. According to the release describing the four papers, capturing that interplay accurately is essential for connecting the two records [3].
The sample is small. The simulations follow one young galaxy on its way to roughly the Milky Way's mass [4], so every gas structure reported comes from a single history. The comparative claim is that simplified models may underestimate how stellar radiation and complex chemistry shape the gas around galaxies [6], and that high resolution picks out structure those models miss [7]. Neither claim says how large the underestimate is, or whether the extra structure changes a quantity a telescope actually records. The release says only that it helps improve predictions for current and future observations [7].
Rey described the payoff as a direct comparison. "MEGATRON allows us to test these ideas directly by comparing detailed simulations with observations from JWST and the chemical fingerprints preserved in ancient stars," he said [11]. The four studies, published in the Open Journal of Astrophysics by Bath with the University of Chicago and the Institut d'Astrophysique de Paris, are the collaboration's first substantial body of results, and more papers are expected [1][2].
For the next round, the project has been awarded 40 million processor hours on the UK's national supercomputers, a sum the release equates to running 5 million laptops in parallel for a year [12]. The two figures do not agree by simple division: 40 million hours spread over a year is about 4,570 processors running continuously [1]. Rey's group at Bath plans to spend it on simulations with higher resolution and more complete physics [13].
What to watch
- Whether later MEGATRON papers publish numerical comparisons between the simulated galaxy and specific JWST measurements or stellar abundance surveys.
- Whether the higher-resolution runs from the 40 million-hour allocation keep finding gas structure that simpler models miss, or whether the results converge.
- Whether anyone puts a size on how much simplified radiation and chemistry shifts predicted observables.