≡ Menu

Atmospheric Turmoil on the Early Earth

Yesterday’s post about planets in red dwarf systems examined the idea that the slow formation rate of these small stars would have a huge impact on planets that are today in their habitable zone. We can come up with mechanisms that might keep a tidally locked planet habitable, but what do we do about the severe effects of water loss and runaway greenhouse events? Keeping such factors in mind plays into how we choose targets — very carefully — for future space telescope missions that will look for exoplanets and study their atmospheres.

But the question of atmospheres on early worlds extends far beyond what happens on M-dwarf planets. At MIT, Hilke Schlichting has been working on what happened to our own Earth’s atmosphere, which was evidently obliterated at least twice since the planet’s formation four billion years ago. In an attempt to find out how such events could occur, Schlichting and colleagues at Caltech and Hebrew University have been modeling the effects of impactors that would have struck the Earth in the same era that the Moon was formed, concluding that it was the effect of small planetesimals rather than any giant impactor that would have destroyed the early atmosphere.

“For sure, we did have all these smaller impactors back then,” Schlichting says. “One small impact cannot get rid of most of the atmosphere, but collectively, they’re much more efficient than giant impacts, and could easily eject all the Earth’s atmosphere.”


Image: An early Earth under bombardment. Credit: NASA.

A single large impact could indeed have dispersed most of the atmosphere, but as the new paper in Icarus shows, impactors of 25 kilometers or less would have had the same effect with a great deal less mass. While a single such impactor would eject atmospheric gas on a line perpendicular to the impactor’s trajectory, its effect would be small. Completely ejecting the atmosphere would call for tens of thousands of small impacts, a description that fits the era 4.5 billion years ago. Calculating atmospheric loss over the range of impactor sizes, the team found that the most efficient impactors are small planetesimals of about 2 kilometers in radius.

Schlichting believes that giant impacts cannot explain the loss of atmosphere, for recent work uses the existence of noble gases like helium-3 inside today’s Earth to argue against the formation of a magma ocean, the consequence of any such giant impact. From the paper:

Recent work suggests that the Earth went through at least two separate periods during which its atmosphere was lost and that later giant impacts did not generate a global magma ocean (Tucker & Mukhopadhyay 2014). Such a scenario is challenging to explain if atmospheric mass loss was a byproduct of giant impacts, because a combination of large impactor masses and large impact velocities is needed to achieve complete atmospheric loss… Furthermore, giant impacts that could accomplish complete atmospheric loss, almost certainly will generate a global magma ocean. Since atmospheric mass loss due to small planetesimal impacts will proceed without generating a global magma ocean they offer a solution to this conundrum.

The same impactors that drove atmospheric loss would, in this scenario, introduce new volatiles as planetesimals melted after impact. By the researchers’ calculations, a significant part of the atmosphere may have been replenished by these tens of thousands of small impactors. What happens to a newly formed planet that can lead to the emergence of life? Learning about the primordial atmosphere shows us a planet at a stage when life was about to take hold, which is why Schlichting’s team wants to move forward with an analysis of the geophysical processes that, in conjunction with small impactors, so shaped Earth’s early environment.

The paper is Schlichting et al., “Atmospheric Mass Loss During Planet Formation: The
Importance of Planetesimal Impacts,” Icarus Vol. 247 (February 2015), pp. 81-94 (abstract / preprint).


Comments on this entry are closed.

  • spaceman December 4, 2014, 15:00


    It seems like there has been (rightfully) a fair amount of investigation into whether or not habitable planets could exist around M-dwarf stars. And, or course, we know from the example of Earth that habitable planets obviously can exist around G-stars. However, I am not aware of too many investigations or modelings of the possibility of habitable planets around K-stars. It would seem to be me that early to mid K- dwarfs might make good habitable planet hosts for a number of reasons. Longer MS lifetime than G-stars, larger HZ than M-stars, fewer flare issues compared to M-stars, and a higher prevalence than G-stars. Also, K-star habitable planets would be further away from their host stars than habitable of M-stars thereby reducing the chances of tidal locking. Perhaps I am just unaware…does anyone happen to know how extensively K-star habitability has been studied?

  • ljk December 4, 2014, 17:10

    Moon’s Long-Ago Magnetic Field May Have Trumped Earth’s

    by Charles Q. Choi, Space.com Contributor | December 04, 2014 02:01
    pm ET

    The moon once had a magnetic heart, which helped the moon generate a magnetic field stronger than Earth’s is now, researchers say.

    However, many mysteries remain about the magnetic field of the moon, such as what powered it and when it ended, the scientists added.

    The moon today does not have a global magnetic field. However, moon rocks that astronauts collected during the Apollo missions suggested the moon once had a magnetic field billions of years ago

    Full article here:


    To quote:

    Much remains unknown about the moon’s magnetic field. “We still don’t know when the lunar dynamo turned off,” Weiss said. “There’s evidence it lasted until at least 3.3 billion years ago, and perhaps as long as 1.3 billion years ago, really pushing the limits of what we know can power the lunar dynamo.”

    Weiss and his colleague Sonia Tikoo reported the latest findings on the lunar dynamo online today (Dec. 4) in the journal Science.

  • Ashley Baldwin December 4, 2014, 17:44

    Maggie Turnbulls list of habitable stars for investigation has a good number of K stars on it . The TESS mission isn’t just about M dwarfs , it also covers later K stars from K4-K9. Trouble is the smaller the star the more prone to flares it is . Even some G stars flare. The cut off for tidal locking within Earths current age is between 0.5 AU and 0.6. That’s about K3 and below are ok . If you type in “The 100 nearest stars to Earth” into Google , one of the first few hits is a list of the hundred brightest stars to Earth and that has all the nearby K stars on it. Quite a few are on Maggie’s list and worth reading about . Omicon Eridani is 20 or so light years away and very interesting. It’s also on the WFIRST / EXO-S list of targets too. Along with a few K dwarfs .

  • Ronald December 5, 2014, 5:58

    This just in:
    Finding infant Earths and potential life just got easier:

  • Ronald December 5, 2014, 6:00
  • Alex Tolley December 5, 2014, 12:14

    I’m not clear from the terms used in this paper whether the Earth originally had a dense atmosphere or not. Was it dense, then stripped and new volatiles added, or was the Earth’s atmosphere in the same range of density as today, and then replaced after stripping?

    I look forward to reading about the importance of this effect on the possible formation of life, if that is possible.

  • Alex Tolley December 10, 2014, 12:09

    And now, courtesy of Curiosity rover, we have evidence that Mars may have had a thick, warm atmosphere for possibly 2 bn years, long after any bombardment may have removed it.

    Curiosity Mars rover ‘solves mountain riddle’