Late, Big Bombardments Brought Heavy Metals to Earth

Article written: 9 Dec , 2010
Updated: 24 Dec , 2015


One of the fundamental problems in planetary science is trying to determine how planetary bodies in the inner solar system formed and evolved. A new computer model suggests that huge objects – some as big as large Kuiper Belt Objects like Pluto and Eris — likely pummeled the Earth, Moon and Mars during the late stages of planetary formation, bringing heavy metals to the planetary surfaces. This model – created by various researchers from across the NASA Lunar Science Institute — surprisingly addresses many different puzzles across the Solar System, such as how Earth could retain metal-loving, elements like gold and platinum found in its mantle, how the interior of the Moon could actually be wet, and the strange distribution in the sizes of asteroids.

“Most of the evidence of what happened during the late stages of planetary formation has been erased over time,” said Bill Bottke from the Southwest Research Institute, who led the research team. “The trail we’ve been tracking on these worlds is pretty cold and to be able to dig more information out of what we have and be able answer some long standing problems is pretty exciting.”

Bottke told Universe Today that the story this new model tells “is not as complicated as it looks at first glance,” he said. “It includes a lot of concepts together, and some of the concepts have actually been around for awhile.”

Bottke and his team have published their results in the journal Science.

The researchers started with the widely accepted theory of how our Moon was created by a giant impact between the early Earth and another Mars-sized planetary body. “This was the most traumatic event the Earth probably ever went through, and that was the time when presumably the Earth and Moon both formed their cores,” Bottke said.

The heavy iron fell to the center of the two bodies, and so-called highly siderophile, or metal-loving, elements such as rhenium, osmium platinum, palladium, and gold should have followed the iron and other metals to the core in the aftermath of the Moon-forming event, leaving the rocky crusts and mantles of these bodies void of these elements.

“These elements love to follow the metal,” Bottke said, “so if the metal is draining to the core, these elements would want to drain with them. So if this is right, what we would expect that rocks derived from our mantle should have almost no highly siderophile elements, maybe 10 to the minus 5th level or so. But surprisingly, that is not what we see. They are only less abundant by a factor of less than 200, compared to what we would expect, a factor of 100,000 or so.”

Bottke said this problem has been argued about since the 1970’s, with various suggestions on how to answer the problem.

“The most viable answer is that after the Moon forming impact took place, there were also other things that hit the Earth during the late stages of planet formation, objects that were smaller, and these smaller objects replenished these elements and gave us the abundance we see today. This is what we refer to as late accretion,” he said.

On the Moon, the same thing was happening. But there was a problem with this scenario. The ratio of these elements on the Earth compared with rocks on the Moon is about 1000 to 1.

“The gravitational cross section of the Earth is about 20 times that of the Moon,” Bottke said, “So for every object that hit the Moon, about twenty should have hit the Earth. And if late accretion delivered these elements, you should have about a 20 to 1 ratio. But that is not what we see—we see a 1000 to 1 ratio.”

Bottke – a planetary dynamacist — discussed this with colleague David Nesvorny, also from SWRI, as well as geophysical-geochemical modelers, such as Richard Walker from the University of Maryland, James Day from the University of Maryland, and Linda Elkins-Tanton from the Massachusetts Institute of Technology.

They came up with a computer model that seemed to provide an answer.

“By playing roulette with these objects, I found that very often the Earth was getting hit by huge impactors that the Moon would never see,” Bottke said. “This result suggests that the things hitting the Earth and Moon at the end of the planet formation period was dominated by very large objects.”

The model predicted that the largest of the late impactors on Earth, at 2,400 – 3,200 km (1,500-2,000 miles) in diameter, while those for the Moon, at approximately 240 – 320 km.

Bottke called that a “cute” result – but they needed more supporting evidence. So, they took a look at the last surviving population of the things that built the planets, the inner asteroid belt. “You find large asteroids like Ceres, Vesta and Pallas” Bottke said, so there are the large ones at 500 to 900 km, but then your next largest asteroids are only about 250 km. This matched up with the sizes that our model came up with,” in which no asteroids with “in-between” sizes are observed in this region.

Maps of Mars' global topography. Credit: NASA

Next, they looked at Mars, which has some very large impact basins which are probably left over from the days of when the planet formed, including the Borealis Basin, which is so large it likely accounts for the differences in the northern and southern hemispheres on the Red Planet.

“We looked and projected the size of the impactors that would have created those impact basins and we saw the distribution of sizes was very much like what was predicted for the Earth and Moon, and also what is found in the inner asteroid belt.

So all those things together — the theoretical basis, the observational evidence from elements on the Earth and Moon, and impacts on Mars collectively says something about the distribution of sizes of objects towards the end of planetary formation.

And what are the implications?

“We could make predictions for what was hitting the Earth, Moon and Mars at that time, and they line up with what we see on the surfaces,” Bottke said. “On Mars we can play a game of what is the biggest projectiles that should have hit Mars, and it matches up well with the size that big basin that formed on Mars, and also produced the abundances of elements we see there.”

“For the Moon, the biggest impactors would be 250-300 km, which is about the size of the south pole Aiken basin,” Bottke continued. “For the Earth, these big impactors explain why some of these impacts managed to hit the Earth and not all the elements went to the core of the Earth.”

Bottke said that adding to the complications, some of the biggest impacts actually may have plowed through the Earth and actually came out the other side — in a very fragmented state — and rained back down on Earth. “If this is true, this provides a way to spread fragments all the way across the Earth,” he said, “but how the debris gets redistributed around the planetary body is a really interesting question. That part needs a lot more work and is simply at the edge right now of what we can do numerically.”

When it comes to water on the Moon’s interior – which was once thought to be dry, but recent sample measurements, however, suggest that the water content in the lunar mantle is between 200 and several thousand parts per billion — Bottke’s model could also address this issue.

“If true,” the team writes in their paper, “it is possible that the same projectile that delivered most of the Moon’s HSEs may have also have provided it with water….Late accretion provides an alternative explanation in case lunar mantle water cannot migrate from the post–giant impact Earth to a growing Moon through a hot and largely vaporized protolunar disk.”

As to why smaller projectiles hit the Moon as compared to Earth, Bottke said it is just a numbers game. “We start with a population which has a certain number of big things, middle sized things and small things,” he said. “And we randomly choose projectiles from that population and for every one big guy that hits the Moon, 20 hit the Earth. And we play that game, and if the number of projectiles is limited, if the Moon only gets hit once or twice from this population, that means the Earth gets hit 20-30 times, that is enough to give us – on most occasions – what we see.”

Bottke said this research gave him a chance to work with geochemists, “who have all sorts of interesting things to say which help constrain the processes that brought about planet formation. The problem is that sometimes they have great information but they don’t have a dynamical process that can work. So by working together I think we were able to come up with some interesting results.”

“The most exciting thing for me is that we should be able to use these abundances that we have on the Earth, Moon and Mars to really tell the story about planet formation,” Bottke said.

Sources: Science, phone interview with Bottke

6 Responses

  1. Astrofiend says

    Certainly seems like they have a strong circumstantial case, which is about all that your likely to get. Nice work on the face of it…

  2. PrometheusOnTheLoose says

    I have often wondered if perhaps there was less gravity on earth during the time of the dinosaurs . . . look how large they are. It just seems to me that they would have a lot of trouble getting around on earth as it is today. The largest land creatures we have now are like bugs compared to the dinosaurs.

    • Uncle Fred says

      They could get around, and gravity was much the same. Thank a generally warmer planet with long eons of relatively stable habitats. It’s only been 65 million years since the KT extinction event and a few thousand since our hunter ancestors decided big beasts make good dinners.

  3. Paul Eaton-Jones says

    “Bottke said that adding to the complications, some of the biggest impacts actually may have plowed through the Earth and actually came out the other side — in a very fragmented state — and rained back down on Earth”.
    Unless this happened when the earth was semi/fully molten surely it would mean that the earth was virtually shattered more than the one time when the Moon was formed. There wouldn’t have been time during the LHB period for the earth to suffer multiple events of this kind, reform, cool, produce an atmosphere and hydrosphere AND produce the organisms that are found just after the LHB. Have I missed something?

  4. Torbjorn Larsson OM says

    This is a finely tuned model on many accounts, relying on today’s asteroid population to boot, and while I haven’t read it yet I don’t vouch for its chances. Why would the fine balance between asteroid sizes, core formation times and impact fragmentation apply in the first place? Why would the steroid population be bimodal? Et cetera. The core formed independently of impactor in the first ~ 30 My last I checked; the Moon impactor is ~ 50 My. (But certainly there is a time period overlap, if you push it.)

    [Disclaimer:] I fancy the new “rapid freeze” model where a fast first crust formation exactly explains the siderophile behavior on Earth and Moon, and where they this year IIRC _predicted_ and retained primordial mantle material from the nowadays Island hot spot on the other side of Greenland. (Can’t remember the island’s name and have no time to check.) It happened to go through a mantle reservoir that survived between first crust and now.

    Such primitive crust formation times would nicely leave enough time for the 4.4 Gy old liquid water results from zircons.

    @ Paul Eaton-Jones:

    I don’t get that either, perhaps it is in the paper.

    My current layman understanding is that you can model bacterial populations surviving in a Goldilocks crustal zone during LHB impacts, and that “crust busters” or rather ocean evaporators merely sterilizes the global surface (if at that) for 100s or thousands of years.

    Now if we are concerned with crust busters, I know that (small scale) hyper velocity impactors leave a crater ~ 20 times their own size. I also learned from the recent findings of deep wet mineral uplifts in the center of a Mars crater that the uplift is empirically fitted by depth ~ 0.1 (0.089, IIRC) of crater size; the ~ 60 km crater had a crater peek with material thrown up from ~ 6 km depth.

    So an ocean crust buster, where AFAIU much of todays ocean plate tectonics surface area has ~ 100 km thickness, needs to open up a crater that is ~ 1000 km to expose mantle material. That crater would come from at least a ~ 50 km diameter impactor. I seem to remember that an ocean evaporator must be at least ~ 300-400 km diameter to impart enough energy to boil it all.

    There are few Moon impacts of crust buster size, none of ocean evaporator size. Given the gravitational cross section ratio of ~ 20 between Earth and Moon, you would see perhaps 10-100 _minimal_ crust busters, likely no ocean evaporator. I don’t think Moon’s Aitken basin would make it, it is too tiny at a mere ~ 2 500 km, suggesting a ~ 120 km impactor. The Moon impactor would though. (O.o).

    The simplest model, suggested by the liquid water test of Jack Hill zircons and the putative biological sulfur cycle suggested by Nuvvuagittuq greenstone belt perhaps dating from before or under LHB*, is that life got started well before LHB. Perhaps LHB is the reason why RNA life disappeared; after all, viruses do well on RNA so why would RNA mesophiles disappear. (It is an iffy hypothesis though, since the LHB bacterial results even allow mesophilic survival – bacteria are so prodigious they are _really_ hard to kill.)

    * And of course possible by the early crust model. Also possible by the latest results on high temp chemical reaction times which simultaneously removes down the need for enzymes, and supply chemical pathways for coenzyme establishment as the temperature lowered down to an RNA world setting. Hmm, I wish I had taken the time to look up and supply the refs now. Well, if prompted to later, certainly.

  5. RUF says

    Large land animals like dinosaurs are not produced by lesser gravity, but by more oxygen. Such large animals needed higher oxygen levels than we have today to survive. Less oxygen = smaller creatures.

Comments are closed.