Scientists Recreate the First 2 Billion Years of the Universe in Stunning New Simulation

MIT and Harvard researchers ran the model on Oak Ridge’s Frontier supercomputer, covering 100 times the volume of earlier simulations

Nikshep Myle Avatar
Nikshep Myle Avatar

By

Simulation of gas structures as electrons were removed from helium by radiation emitted by early stars and black holes. Oliver Zier / Center for Astrophysics | Harvard & Smithsonian

Key Takeaways

Key Takeaways

  • Lumina simulates 1.6 billion light-years of early cosmic history using half a trillion computational elements.
  • Researchers use Lumina to translate reionization theories into predictions testable against James Webb Space Telescope data.
  • Lumina covers more than 100 times the volume of comparable simulations, capturing rare early galaxy environments.

Half a trillion computational elements, thousands of GPUs, and one very old question: how did a dark, neutral universe become filled with galaxies and light? That question drove researchers from MIT, the Center for Astrophysics | Harvard & Smithsonian, and partner institutions to build Lumina. The radiation-hydrodynamics simulation covers approximately 1.6 billion light-years across the first 2 billion years of cosmic history.

The Dark Ages Were Exactly That

Neutral hydrogen once made the young universe opaque, and understanding how that changed remains one of cosmology’s central unsolved problems.

After the Big Bang, neutral hydrogen filled intergalactic space and made it opaque to ultraviolet radiation. The first stars and galaxies eventually emitted enough ultraviolet light to strip electrons from that hydrogen, a process called reionization, allowing radiation to travel more freely through intergalactic space.

A patchy map of hydrogen ionization, which happened mainly around stars. Oliver Zier / Center for Astrophysics | Harvard & Smithsonian

What Lumina Actually Models

The simulation tracks every major component of early cosmic structure, starting about 50 million years after the Big Bang.

Lumina begins about 50 million years after the Big Bang and tracks the coupled evolution of dark matter, gas, stars, galaxies, black holes, and radiation through the reionization transition. According to the Smithsonian Institution, it is the first simulation of its type to model helium reionization and let those effects feed back into the developing universe within the same calculation.

The Hardware Behind It

Frontier, the supercomputer at Oak Ridge National Laboratory, provided the computing power the project required.

Lumina ran on the Frontier supercomputer at Oak Ridge National Laboratory. The model contains approximately half a trillion computational sectors and covers more than 100 times the volume of comparable earlier simulations while retaining enough resolution to study individual developing galaxies.

Helium reionization. Blue indicates regions being ionized by supermassive black holes. Oliver Zier / Center for Astrophysics | Harvard & Smithsonian

Translating Theory Into Predictions

Scale matters because smaller simulations can miss the rare structures and varied environments where early galaxies formed.

Smaller simulations can miss rare large-scale structures or fail to represent the full range of environments where early galaxies formed. Oliver Zier, lead author and computational astrophysicist at the Center for Astrophysics | Harvard & Smithsonian, stated the core purpose plainly:

“To understand what we are seeing with [these telescopes], we need to translate our theories into detailed predictions.” Oliver Zier, lead author and computational astrophysicist, Center for Astrophysics | Harvard & Smithsonian, via Smithsonian magazine.

Where the James Webb Telescope Enters

Lumina generates predictions that researchers can compare against James Webb Space Telescope observations of the distant universe.

Lumina’s predictions can be compared with observations from NASA’s James Webb Space Telescope, which has been identifying increasingly distant galaxies since its first science data arrived in 2022. When simulated galaxy counts, brightnesses, and radiation patterns match what JWST observes, that agreement supports the model’s underlying assumptions. Where they diverge, researchers gain a specific target: the discrepancy points to missing physics or prescriptions for star formation and black-hole growth that need revision.

A cosmic map showing the distribution of ordinary matter, which makes up the stuff we can observe, and dark matter, a mysterious, invisible substance that acts as cosmic glue. Red regions show locations with a higher ratio of ordinary matter to dark matter, and arrows show the velocity of the gas Oliver Zier / Center for Astrophysics | Harvard & Smithsonian

A Framework, Not a History

Lumina is a testable theoretical tool, and its results are shaped by the assumptions and approximations built into it.

A simulation depends on the physical laws and formation rules built into it, not to mention its initial conditions and numerical methods. Lumina is a testable theoretical framework, not a confirmed account of what happened.

Why It Becomes a Reference Point

As JWST pushes deeper into cosmic history, Lumina provides a large-scale prediction to test against what the telescope finds.

As JWST continues identifying distant galaxies, Lumina gives researchers a high-resolution prediction to test against those observations. The model could serve as a reference dataset for studying first-galaxy formation, early black-hole growth, and the structure of the intergalactic medium between those ancient galaxies.

Share this

At Gadget Review, our guides, reviews, and news are driven by thorough human expertise and use our Trust Rating system and the True Score. AI assists in refining our editorial process, ensuring that every article is engaging, clear and succinct. See how we write our content here →