Ultra-faint dwarf galaxies, tiny satellite galaxies orbiting the Milky Way, have long been seen as cosmic fossils. Now, a new study by researchers at the Oskar Klein Centre and the LYRA collaboration uses an unprecedented set of simulations to show just how powerfully these faint systems can reflect the conditions of the early Universe and tell us why some galaxies grew and others did not.
Azadeh Fattahi is Associate Professor at the Oskar Klein Centre (OKC) and heading the research group which led this work, now published in Monthly Notices of the Royal Astronomical Society (MNRAS), together with collaborators from Durham University and University of Hawaii. She explains the scale of the project:
"In this work we presented a brand-new suite of cosmological simulations focused on the faintest galaxies in the Universe, with an unprecedented resolution. These are by far the largest sample of such galaxies ever simulated at these resolutions."
Dwarf galaxies are often described as small cousins of the Milky Way. They form in small dark matter halos which are predicted by the standard model of cosmology. The faintest examples of such systems are extreme in both size and fragility, and lie on the boundary of our knowledge about galaxy formation and dark matter.
"The smallest galaxies are called ultra-faint dwarf galaxies, which are a million times less massive than the Milky Way or even smaller," Fattahi says. "Due to their small size these galaxies have proven very difficult to model and simulate."
This new simulation suite represents a major step forward, enabling a systematic view of how these galaxies form and evolve.
A down-to-earth analogy
"A useful analogy is to plants and crops and how the way they grow is sensitive to the weather conditions", says Shaun Brown who led the study while working at OKC and Durham University. "In the same way that the yield of a crop in summer can indirectly tell you a lot about what the weather in spring must have been like, the properties of faint dwarf galaxies today can tell us a lot about the conditions, or weather, of the Universe at a much earlier time."
What makes the results especially timely is that the simulations do more than reproduce faint dwarf galaxies – they suggest that these local objects can act as a probe of the Universe's earliest "climate". The team explored how different assumptions about the early radiation environment influence which small dark matter haloes manage to form stars at all.
"In the paper we studied two different assumptions about the properties of the early Universe when it was less than 500 million years old, to understand the effect on the properties of these small galaxies today when the Universe is 13 billion years old," Brown explains.
The outcome was striking:
"We found that these small ultra-faint galaxies are very sensitive to these changes, while more massive galaxies, like our Milky Way, don't really care," he adds, "For the smallest galaxies, early conditions can decide whether they become visible galaxies – or remain starless dark matter halos."
Future research
That sensitivity opens a clear path to testing early-Universe physics with upcoming observations.
"Excitingly, in the near future we will have data from the Vera C. Rubin Observatory which will be able to find many more of these ultra-faint dwarfs around the Milky Way," says Fattahi.
Many astronomers hope the Vera C. Rubin Observatory can deliver a near-complete census of Milky Way satellite galaxies – and these simulations suggest that census may carry information far beyond our local neighbourhood.
"Our work suggests that these upcoming observations of the very local Universe will be able to constrain what the Universe at its infancy looked like, something we currently cannot directly access with other observations."
The result is particularly relevant in light of recent discoveries of galaxies in the early Universe, by the James Webb Space Telescope (JWST), which is finding many surprises, in particular unexpectedly massive and bright galaxies in the early universe," Fattahi notes.
If the early Universe is producing surprises at large distances, then local relics from the same epoch, ultra-faint dwarfs, may provide an additional route to understanding what happened.
Reaching this regime came with major practical challenges. "Running these simulations is challenging, and extremely expensive in both time and computational resources," Fattahi says.
In total it took more than six months to run all of the simulations. The scale of the data was also substantial: "The simulation also produces very large amounts of data (in total approximately 300 TB). This meant many of the old algorithms designed for smaller amounts of data needed updating and improving to effectively handle this new large amount of data."
Work carried out on the COSMA 8 supercomputer
Most of the work was carried out on the COSMA 8 supercomputer, which is designed for simulation-driven research. Durham University's Institute for Computational Cosmology hosts COSMA 8 on behalf of the UK's DiRAC High Performance Computing Facility.
Looking ahead, Fattahi's team plans to use the new suite to tackle questions that are still open in modern galaxy and structure formation, such as where we can find the very first generation of stars formed in the Universe or what do the properties of ultra-faint dwarf galaxies tell us about the nature of dark matter?
Read article in Monthly Notices of the Royal Astronomical Society (MNRAS)
