milky way

On the left the halo appears messy and ‘wrinkly’, a sign that a merger has occurred relatively recently. On the right it appears smooth and uniform, a sign that a merger has instead occurred in the ancient past. Credit: Halo stars: ESA/Gaia/DPAC, T Donlon et al. 2024; Background Milky Way and Magellanic Clouds: Stefan Payne-Wardenaar;; LICENCE: CC BY-SA 3.0 IGO or ESA Standard License

‘Benjamin Button’ effect allows astronomers to better calculate ‘last big crash’

TROY, N.Y. — A stunning new study may upend everything we know about our cosmic home — the Milky Way galaxy. According to researchers from the Rensselaer Polytechnic Institute, our galaxy may have collided with another galaxy billions of years later than scientists previously thought. In fact, according to the study, the last time the Milky Way collided with another star cluster, the Earth had already formed. What a light show that must have been!

The findings, in a nutshell

In their groundbreaking study, published in the Monthly Notices of the Royal Astronomical Society, the team of astronomers uncovered compelling evidence that the Milky Way galaxy experienced a massive merger event with a dwarf galaxy about six billion years later than believed. This discovery challenges the long-held theory that the last major merger, known as the Gaia-Sausage/Enceladus (GSE), occurred an astounding eight to 11 billion years ago. Instead, the new research suggests that the debris we see in the Milky Way’s stellar halo — the diffuse sphere of stars surrounding the galaxy’s disk — is the result of a collision that took place a mere one to two billion years ago, a cosmic blink of an eye in astronomical terms.

Researchers Heidi Jo Newberg and Tom Donlon focused on the “wrinkles” in our galaxy, which form when other galaxies smash into the Milky Way.

“We get wrinklier as we age, but our work reveals that the opposite is true for the Milky Way. It’s a sort of cosmic Benjamin Button, getting less wrinkly over time,” says Donlon, lead author of the new Gaia study, in a media release. “By looking at how these wrinkles dissipate over time, we can trace when the Milky Way experienced its last big crash – and it turns out this happened billions of years later than we thought.”

“For the wrinkles of stars to be as obvious as they appear in Gaia data, they must have joined us no less than three billion years ago – at least five billion years later than was previously thought,” adds Newberg. “New wrinkles of stars form each time the stars swing back and forth through the center of the Milky Way. If they’d joined us eight billion years ago, there would be so many wrinkles right next to each other that we would no longer see them as separate features.”

Planet Earth in front of the Milky Way galaxy
According to the study, the last time the Milky Way collided with another star cluster, the Earth had already formed. (Credit: muratart/Shutterstock)

Methodology: Decoding the Clues

To unravel this mystery, the researchers employed a variety of cutting-edge techniques. First, they developed a semi-analytical model that relates the number of “caustics” (wrinkles or folds in the phase-space distribution of stars) to the time since a merger event. By analyzing data from the Gaia space observatory, the team identified several caustics in the local stellar halo and used their model to estimate the time of this collision.

However, the team didn’t stop there. They delved deeper into the dynamics of these caustics by comparing their observations to a state-of-the-art cosmological simulation of a Milky Way-like galaxy. This simulation, part of the FIRE-2 Latte suite, allowed them to track the evolution of a simulated dwarf galaxy as it crashed into the host galaxy at different times.

To make the comparison as accurate as possible, the researchers introduced a novel metric called “causticality,” which calculates the degree of unevenness in the phase-space distribution of stars. A high causticality value shows that the stars are not yet fully phase-mixed, suggesting a more recent collision.

Key Results: A Cosmic Collision in Recent History

The results of this analysis were nothing short of astonishing. The observed data from the Gaia observatory exhibited a high causticality value, revealing the presence of prominent, asymmetric caustics. When compared to the simulated data, the observed causticality matched the simulated merger debris best at a time roughly one to two billion years after the collision.

This finding stands in stark contrast to the widely accepted scenario of the GSE merger, which is thought to have occurred between eight and 11 billion years ago — long before Earth formed. The researchers found that in these ancient cosmic times, the simulated data showed a much lower causticality, pointing to a higher degree of phase-mixing than what is observed in the Milky Way’s stellar halo today.

Study Limitations

While the study presents compelling evidence for a recent merger event, the researchers acknowledge several limitations and challenges. One key limitation is the reliance on a single cosmological simulation, which may not fully capture the complexities of the Milky Way’s formation history.

Additionally, the observed data is limited to the local stellar halo within five kiloparsecs (about 16,000 light-years) of the Sun. It is possible that the phase-space distribution of stars at larger distances could reveal a different picture.

Another challenge lies in the up-sampling process used to increase the resolution of the simulated data. While necessary for a meaningful comparison, this process could introduce biases or underestimate the true degree of phase-mixing.


Despite these limitations, the researchers argue that their findings are robust and consistent with other lines of galactic evidence. For instance, the observed stellar shells and substructures in the Milky Way’s halo are better explained by a recent collision rather than an ancient one, as older debris would have had more time to phase-mix and become less pronounced.

Moreover, the study provides a compelling alternative to the GSE scenario, which has faced increasing scrutiny in recent years. Some researchers have argued that the chemical and kinematic signatures attributed to the GSE could be explained by other processes, such as secular evolution or multiple smaller mergers.

If confirmed, this study’s findings could profoundly impact our understanding of the Milky Way’s formation history and the role of mergers in shaping galaxies. They could also have implications for our knowledge of galaxy evolution in general, as the timescales and dynamics of mergers are crucial for modeling and interpreting observations.

“Through this study, Doctors Newberg and Donlon have made a startling discovery about the history of the Milky Way galaxy,” says Curt Breneman, Ph.D., dean of Rensselaer’s School of Science.

About Chris Melore

Chris Melore has been a writer, researcher, editor, and producer in the New York-area since 2006. He won a local Emmy award for his work in sports television in 2011.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink


Chris Melore


Sophia Naughton

Associate Editor


  1. Ronald L Giuntini says:

    It would NOT have been a “LIGHT SHOW!!”.
    Galaxies in collision to not involve crashing planets and crashing stars.
    The galaxies move into one another, and gravitational effects induce distortion of the individual galaxies, but there is no violence involved.

    1. Bob Battle says:

      I don’t believe the article indicated a violent collision type light show. You may have taken the word “collision” that way. I would think taking a look at the Milky Way as we know it, and seeing 2 close up galaxies instead of one would indeed be quite a light show.

      1. Bill Wedekind says:

        Quoting the first paragraph of the article:

        “What a light show that must have been!”

        So, there was an assertion of a “light show” in the article. You may not have read the article.

    2. Peter Limput says:

      What about the violence between the astro moon men and the martian crab people?

  2. Edward Smith says:

    If the Earth were as it is today, atmosphere, ocean coverage, forestation, climate, etc. What would it look like after the merger?

  3. Goatse McGoatface says:


    1. GiraffeSweaters says:

      Getting upset and calling people names just because you don’t have the capacity to understand what’s going on is very childish.

    2. Walter Ziobro says:

      That comment leaves me flat.

  4. Eric says:

    One or two billion years is a cosmic blink of an eye?? I mean at some point not everything is a cosmic “blink of an eye”. 1-2 billion out of 13.7 billion years, scaled to a 75 year human lifespan is between 5.5 to 11 years. What is with the stupid obsession with calling any time period a “cosmic blink of an eye”? Who wrote this article, chatgpt?

  5. Liliana says:

    Has anyone ever wonder what is behind all this ? Why not ask the right question instead of análisis. The answers have always been here. Everything repeats itself. Learn to see what you look at. What is in the human being repeats itself outside. But most of all ask yourselves who or what is behind.

  6. Alexandron Janoulis says:

    When I read articles such as this it makes me realize humans can certainly improve our terrestrial environment, but when it comes to climate and far beyond it’s way above human pay grade.

  7. Brian says:

    Yet again we have to rethink “everything we know.” How many times is this now?!?!

  8. Renato Anzalone says:

    ‘a mere one to two billion years ago, a cosmic blink of an eye in astronomical terms.’

    Let’s talk In a trillion years. Then I’ll take that comment serious.

    Regardless, why can’t we have had more than one merger during the last six billion years.

  9. Nancy Ruth O'Hara says:

    I wonder what these findings mean to a secular (not in the religious way) populace, whose lives can hardly be impacted by the knowledge. Certainly, while it means a lot to scientists and others who live on their ability to know, it has a small chance of meaning anything to us who are challenged to understand the miniscule events of our own lives. No disrespect intended — just an observation.

  10. Mystick says:

    This is not really new information. Sensationalist writing.

    1. GiraffeSweaters says:

      It can be new information to someone that didn’t know about it. Relax chief.

  11. Joseph L Snyder says:

    Newberg is quoted as saying: For the wrinkles of stars to be as obvious as they appear in Gaia data, they must have joined us no less than three billion years ago.

    Doesn’t she mean “no more than”?

  12. Alvin Ellis says:

    One clear winter night in 1969 I was in a training exercise near Norman, OK, with the stars so bright we could read the larger details of our maps. No flashlights, headlights, etc. – just starlight, brilliant starlight. And these guys keep changing their minds and saying “This is fact, no. now this, no now…”
    And what continues to baffle them is only a byline in the real creation story, “He made the stars, too.” They’re made with the same maturity as the rest of His creation, not billions of years ago but only a few thousands.
    If want to know what happened back then don’t look up to see it, look down at Genesis in the Bible. And maybe, playfully, you’ll find that Creator Who loves you and me so much He’s sent His Son, Jesus, to be our propitiation and mediator between us and Him.

  13. Alvin Ellis says:

    Don’t you just hate that autocorrect thing! Itomits some yous and changes prayfully to playfully.

  14. Michael says:

    In time Andromeda will devour the Milky Way.
    Galaxies have to eat y’know.
    In heaven above as on earth below.

  15. Nik says:

    Soooo, scientists were 6 billion years off? Does that shake the confidence of the public about other things that they are solid on?

  16. Walter Ziobro says:

    “Additionally, the observed data is limited to the local stellar halo within five kiloparsecs (about 16,000 light-years) of the Sun. It is possible that the phase-space distribution of stars at larger distances could reveal a different picture.”

    An invitation for someone to build a bigger model of the Milky Way.

  17. James says:

    I’m going to order a Gaia Sausage Enceladus for lunch today.