Late Heavy Bombardment

• 1. Introduction

  ›

  The Late Heavy Bombardment describes a period roughly 4.1 to 3.8 billion years ago when something went catastrophically wrong in the inner Solar System. Asteroids and comets rained down on Mercury, Venus, Earth, the Moon, and Mars in numbers that dwarf anything the planets have experienced since. The hypothesis calls it a "lunar cataclysm" for good reason: if the cratering rates measured on the Moon are scaled to Earth, the numbers are staggering. Tens of thousands of craters larger than twenty kilometres across, dozens of basins stretching a thousand kilometres wide, and several impacts large enough to gouge basins five thousand kilometres in diameter. Whatever caused this event did not leave Earth unchanged.

  The evidence for it sits in a cabinet in Houston. Moon rocks collected by the Apollo 15, 16, and 17 astronauts came back from landing sites chosen for their proximity to three of the Moon's great impact basins: Imbrium, Nectaris, and Serenitatis. When scientists dated the impact-melt rocks within those samples, they found that the rocks had last been liquefied by collisions within a remarkably narrow window of time. That clustering was not expected. It suggested a pulse, not a gradual decline. And if the Moon was being hammered that hard, Earth was too. The question that haunts every chapter of this story is whether the Late Heavy Bombardment was the crucible in which life either died, survived, or was first born.

• 2. Reading the Moon Rocks

  ›

  Apollo 15, 16, and 17 returned samples from sites deliberately positioned near three of the largest impact basins on the Moon's nearside. Scientists applied radiometric dating to the impact-melt rocks embedded in those samples, and the results pointed to a cluster of ages falling between roughly 3.8 and 4.1 billion years ago. The investigators called this the "lunar cataclysm" and proposed that it marked a sudden spike in the bombardment rate around 3.9 billion years ago.

  Lunar meteorites added a second line of evidence. These rocks are thought to randomly sample the lunar surface, including regions far from any Apollo landing site. Many feldspathic lunar meteorites probably originated on the lunar far side. When scientists dated impact melts within those meteorites, none turned out to be older than about 3.9 billion years, which is consistent with the cataclysm hypothesis. The ages did not cluster tightly at that date, however; they spread between 2.5 and 3.9 billion years ago.

  Meteorites from the asteroid belt broadened the picture further. Howardite, eucrite, and diogenite meteorites, along with H chondrite meteorites, show numerous ages in the 3.4-4.1 billion year range, alongside an earlier peak at 4.5 billion years ago. Computer simulations using hydrocode revealed a striking physical reason: as impact velocity increases from the current asteroid belt average of five kilometres per second to ten kilometres per second, the volume of impact melt produced rises by a factor of a hundred to a thousand. Those high velocities require the large eccentricities found on planet-crossing orbits, which are rare today but would have been far more common if the giant planets had once disrupted the asteroid belt through orbital migration.

• 3. Doubts and Dissent

  ›

  Not every planetary scientist accepts that the spike in impact-melt ages represents a real bombardment spike. The first major objection zeroes in on the Apollo landing sites themselves. All three sites sit on the near side of the Moon, and quantitative modelling shows that significant amounts of ejecta from the Imbrium impact basin, the youngest and largest of the multi-ring basins on the central nearside, should be present at all three locations. Under this reading, the apparent cluster of ages near 3.9 billion years reflects material scattered from a single enormous impact, not a swarm of independent events spanning tens of millions of years.

  A second objection questions what the absence of older impact-melt rocks actually means. The standard interpretation treats this gap as evidence that few large impacts occurred before 4.1 billion years ago. Critics counter that old melt rocks may have existed but had their radiometric ages reset by four billion years of continuous cratering. Alternatively, those samples may have been pulverised to sizes too small for standard radiometric methods to yield reliable dates. The 40Ar/39Ar dating system, which detected the age spike at 3.9 billion years, could also produce a similar pattern if the early crust formed episodically and then lost argon gradually as the impact rate declined rather than in a single catastrophic pulse.

  One emerging alternative reframes the entire event. Rather than a brief cataclysm, a range of evidence points toward a more extended period of lunar bombardment lasting from approximately 4.2 billion years ago to 3.5 billion years ago. That is a span of roughly 700 million years, far longer than the cataclysm model requires. Matija Cuk proposed a different mechanism altogether: that a Vesta-sized Mars-crossing asteroid, a remnant of an originally much larger population, was catastrophically disrupted roughly 3.9 billion years ago, flooding the inner Solar System with debris that produced the last few lunar impact basins.

• 4. Possible Causes

  ›

  The Nice model, named after the French city where it was developed, became the dominant explanation among planetary scientists. In its original form, the simulations by Gomes and colleagues began with the giant planets packed into a tight orbital configuration surrounded by a rich trans-Neptunian belt. Objects from that belt slowly drift onto planet-crossing orbits, nudging the giant planets to migrate over several hundred million years. When Jupiter and Saturn's orbits drift apart far enough to cross a 2:1 orbital resonance, their eccentricities jump. Uranus and Neptune scatter outward, disrupting the outer belt and sending comets into the inner Solar System. Resonances sweeping through the asteroid belt simultaneously kick many asteroids onto Earth-crossing trajectories.

  The model has since been revised. In the updated version, the giant planets start in a multi-resonant configuration produced by early gas-driven migration through the protoplanetary disk. A "jumping-Jupiter" scenario replaces the slow resonance drift: an ice giant encounters Saturn, is propelled onto a Jupiter-crossing orbit, and is then flung outward by Jupiter. This rapid separation limits the damage done to the asteroid belt and helps preserve the low eccentricities of the terrestrial planets. One consequence is that the inner asteroid belt, now nearly depleted, becomes the primary impactor source. Some recent work, however, finds that the inner asteroid belt alone cannot account for the ancient impact spherule beds and the large lunar basins, leaving the source of the bombardment genuinely unsettled.

  A separate hypothesis centres on a hypothetical fifth terrestrial planet called Planet V. With a mass less than half that of Mars, Planet V originally orbited between Mars and the asteroid belt. Perturbations from the other inner planets eventually drove its orbit into the inner asteroid belt, destabilising many asteroids and sending them onto Earth-crossing paths. Planet V was ultimately lost, most likely by falling into the Sun. An alternate version of this scenario proposes that Planet V struck Mars, forming the Borealis Basin, and that the impactors responsible for the lunar basins were the debris from that collision.

• 5. Earth's Erased Record

  ›

  If the cataclysm was real, Earth absorbed the same punishment as the Moon, amplified by its larger surface area and gravitational pull. Before the LHB hypothesis existed, geologists assumed that Earth had remained mostly molten until about 3.8 billion years ago. The oldest known rocks from around the world seemed to converge on that date, and uranium-lead dating of zircons, considered the most accurate and environmentally stable method available, appeared to confirm a hard cutoff. The boundary between the earlier Hadean and later Archean eons was drawn there.

  Then, in 1999, the oldest known rock on Earth was dated to 4.031 plus or minus 0.003 billion years old. It belongs to the Acasta Gneiss, found within the Slave Craton of northwestern Canada. Even older individual minerals survive: a zircon from the Jack Hills portion of the Narryer Gneiss Terrane in Western Australia has been dated to 4.404 billion years. That zircon likely represents a fragment of crust that predates the LHB, now preserved inside a much younger rock roughly 3.8 billion years old.

  The Jack Hills zircon forced a rethink of what the Hadean eon actually looked like. Older textbooks depicted a hellish molten surface covered in volcanoes, a picture that gave the eon its name from the Greek word for the underworld. Isotopic ratios preserved in zircon crystals, however, hinted at a very different reality: a solid surface, temperate conditions, and acidic oceans. Whether that picture is accurate remains controversial, but it shifted the default assumption away from a featureless magma ocean and toward a geologically complex early Earth interrupted by the bombardment rather than created by it.

• 6. Life at the Edge

  ›

  Manfred Schidlowski argued in 1979 that carbon isotopic ratios in certain sedimentary rocks found in Greenland were a relic of organic matter. The ratio of carbon-12 to carbon-13 in those rocks was unusually high, which is normally a sign of biological processing. Schidlowski dated the rocks to about 3.8 billion years ago; others suggested a more modest 3.6 billion years. Either date leaves an uncomfortably short window for abiogenesis to have produced living organisms, and if Schidlowski's older date was correct, it was arguably too short.

  A 2002 study supported the older end of that age range, placing the Greenland rocks at about 3.85 billion years ago. If the Late Heavy Bombardment indeed sterilised Earth's surface, the implication is striking: life must have either survived the bombardment in sheltered environments or originated in the relatively brief window immediately after it ended. Three-dimensional computer models developed in May 2009 by a team at the University of Colorado at Boulder suggested a third possibility. Their simulations indicated that while the bombardment would have sterilised Earth's surface, hydrothermal vents below the ocean floor could have sheltered thermophile microbes throughout the event.

  Studies published in 2005, 2006, and 2009 found no support for the isotopically light carbon ratios that formed the original basis of Schidlowski's claim. A 2008 study of Jack Hills rocks, however, found traces of similar potential organic indicators. Thorsten Geisler of the Institute for Mineralogy at the University of Munster examined carbon trapped in tiny pieces of diamond and graphite inside zircons dated to 4.25 billion years, pushing the tentative signs of organic chemistry back even further. In April 2014, scientists reported finding evidence of the largest known terrestrial meteor impact near the Barberton Greenstone Belt, estimating it struck about 3.26 billion years ago with an impactor roughly 37 to 58 kilometres wide, a reminder that the bombardment's legacy on Earth remains only partially read.