Science News logo

Science Service logo


Water for the Rock
Did Earth's oceans come from the heavens?

Ben Harder

More than 4.5 billion years ago, the sun and its planets were taking shape from a rotating disk of ice, gas, and dust. This protosolar nebula was hotter and denser toward its center and cooler and less dense farther out. These gradients profoundly influenced the chemical composition of different regions of the early solar system, including the distribution of water. Close to the nebula's center, high temperatures and pressures vaporized ice crystals and the light elements and compounds called volatiles. The action blew these materials toward the outskirts of the nebula, leaving mainly grains of rock behind to form the inner planets.

photo

CELESTIAL RAIN. Toward the end of Earth's formation, a steady bombardment of ice-containing meteorites could have provided much of the planet's water.

M. A. Garlick / Space-art.co.uk

Farther out, debris coalesced in meteorites called carbonaceous chondrites, which carry up to 10 percent of their mass in ice. The giant outer planets, such as Saturn and Jupiter, that arose in this neighborhood also contain some ice. Beyond these planets, water condensed in large quantities and formed comets, which are about half ice.

Compared with these icy objects, Earth contains little water. Only about 0.02 percent of its mass is in its oceans, and somewhat more water sits beneath the surface. Nevertheless, Earth has substantially more water than scientists would expect to find at a mere 93 million miles from the sun. How did Earth come to possess its seas?

Over the years, planetary scientists have proposed several possible answers to that question, but until recently they've had little data for testing their hypotheses. As research in the field progresses, however, the picture is getting more complicated—not less.

Analyses of the geochemical properties of various bodies in the solar system and computer modeling of the dynamics of ancient planetary interactions have undermined a formerly popular theory, which attributes Earth's water to a bombardment by comets late in the planet's formation.

New hypotheses are emerging as that theory's plausibility fades, and planetary scientists are struggling to reconcile data with these alternative scenarios. There's one thing on which most geochemists and astronomers agree: The celestial pantry is now empty of a key ingredient in the recipe for Earth.

Just add water

Because comets contain a greater proportion of water than other known celestial objects do, they make natural candidates as a source of Earth's rivers, lakes, and oceans. The distribution of hydrogen and water beneath Earth's surface suggests to many geochemists that water hasn't mixed deep into the planet, so they thought that the cometary bombardment applied a veneer of water to the dry planet relatively late in its formative period.

One attraction of this late-veneer scenario has been that it fits well with the early movements of planets and the many comets in the outer solar system, says Armand H. Delsemme, an astrophysicist now retired from the University of Toledo in Ohio. As Jupiter formed, its growing gravitational tug would have sent many icy comets hurtling from the range of the giant planets to all reaches of the solar system.

Over a billion years, at least hundreds of millions of comets collided with Earth, Delsemme says. The bombardment would have been especially heavy just after Earth formed.

Attributing water on Earth to these latecomer comets neatly explains a couple of things: first, how water that originated at the outer edges of the solar system got to at least one of its inner planets, and second, how water arrived late enough in Earth's formation for the planet to have sufficient gravity to retain it.

"The front-runner [hypothesis] until about 5 years ago was that water came from comets and came in late," says Kevin Righter, a planetary geochemist at the University of Arizona in Tucson. "One group of measurements changed that."

Those measurements were spectral analyses of the chemical compositions of three comets—Halley, Hyakutake, and Hale-Bopp—during near-Earth passes they made in 1986, 1996, and 1997, respectively. These analyses, the first that examined the hydrogen in water on bodies from a remote region, revealed a crucial chemical difference between the hydrogen in cometary ice and that in Earth's water.

Most hydrogen atoms possess a nucleus made up of a sole proton. Rarer forms also contain a neutron or two. The one-proton- one-neutron version, called deuterium, behaves chemically like hydrogen and can form water and other compounds. However, the resulting molecules are distinctly heavier than those containing the more common form, or isotope, of hydrogen.

Deuterium is exceedingly rare on Earth. Barely one such isotope exists for every 7,000 atoms of standard hydrogen. In contrast, the deuterium-to-hydrogen ratios in the three comets, according to the new observations, were all twice that in Earth's water.

The discovery gave researchers some pause. Assuming that the compositions of Halley, Hyakutake, and Hale-Bopp are representative of all comets, explaining how a hail of the objects could produce oceans with an earthly deuterium-to-hydrogen ratio is like trying to make a low-fat dessert from heavy cream.

According to the new data, cometary bombardment could account for no more than half of Earth's inventory of water, says Francois Robert, a geochemist at the Museum of Natural History in Paris and one of several researchers who brought the paradox of the incompatible ratios to light.

Such numbers might still fit a revised version of the late-veneer theory, says Leonid M. Ozernoy of George Mason University in Fairfax, Va. In addition to comets, asteroid-size planetesimals containing water with less deuterium could have contributed to the late veneer, says Ozernoy.

Smaller versions of these meteorites, the carbonaceous chondrites, hit Earth today in modest numbers. According to a computer model Ozernoy and his George Mason University colleague Sergei Ipatov have built, greater quantities and larger chunks of such material could have showered Earth toward the end of its formation.

Ozernoy and Ipatov have estimated the number of planetesimals that were flung at the early Earth from reservoirs of such bodies following orbits inside Jupiter's path or crossing it. These planetesimals could have delivered much of Earth's water, Ozernoy argued in January at the American Astronomical Society meeting in Washington, D.C.

Wet birth

Adding wet planetesimals to the equation of Earth's early years puts a different face on the late-veneer theory, but it still doesn't satisfy many of the geochemical constraints that have been recently described, says Tobias C. Owen of the University of Hawaii in Honolulu.

Water isn't the only matter on our planet today that seems unlikely to have formed at Earth's proximity to the sun. There are also compounds and elements that readily vaporize, including chemically inert noble gases, such as argon, krypton, and xenon, and the elements nitrogen, oxygen, and hydrogen.

The ratio of xenon to krypton differs between Earth's atmosphere and typical carbonaceous chondrites today. By the same token, the argon-to-water ratios are dissimilar. Therefore, these wet meteors' larger kin, the planetesimals, probably didn't provide a veneer of material for Earth, Owen's analysis suggests.

The isotope profiles of nitrogen and oxygen on meteorites and Earth also argue against these bodies providing much of a wet veneer.

Michael J. Drake of the University of Arizona, who works with Righter, agrees that a late veneer didn't provide Earth's water. While he and Righter don't dispute that a veneer accounts for some of Earth's material, it couldn't have been wet. Certain metals, such as osmium, would have been pulled into Earth's central core if they had been present before the planet got wet. Therefore, all osmium in Earth's upper layers must have come in as a late veneer

Drake and Righter have determined that the isotope profile of near-surface osmium closely matches that in ordinary chondrites—a type of meteorite that's bone-dry. And since carbonaceous chondrites don't have the right proportion of osmium isotopes, they couldn't have made a substantial contribution to the late veneer, the researchers note in the March 7 Nature.

Taken together with the signatures of volatiles on Earth, these data suggest that no more than 50 percent, and probably less than 15 percent, of Earth's water could have been added from space at the end of our planet's formation, says Drake.

If existing objects in space couldn't have combined to make Earth's unique mix of water and other elements, the planet must have formed from—and entirely depleted—an ancient supply of water-rich material that has no modern analog, Drake and Righter argue. Because their hypothesis requires that Earth arose from water-containing materials already present in the inner solar system, it's called the wet-accretion hypothesis.

"Most of Earth's water has an indigenous origin," says Drake. The most probable source is a water-containing inner solar system reservoir at about the same distance away from the sun as Earth is now.

In the wet-accretion hypothesis, Earth developed from silicate rocks with water trapped inside. This hydrous material coalesced with other objects occupying the same swath of space. In their Nature report, Drake and Righter suggest that the band of the solar nebula was cooler than the temperature other researchers have inferred, thus allowing water ice to condense and become bound to the silicates.

One big splash

The role of chance in the solar system's evolution represents a wildcard that could trump both the late-veneer and wet-accretion models. Or it could fold for lack of hard evidence.

Allessandro Morbidelli of the Observatory of the Côte d'Azur in Nice, France, accepts Drake and Righter's hypothesis that Earth formed wet. However, he doubts that the planet evolved solely from material within a tight band at a specific distance from the sun, as the Arizona researchers envision. Their scenario isn't consistent with computer simulations of planetary formation, he says.

Morbidelli returns to the notion that bodies from the outer solar system brought water and volatiles to the inner solar system, but he hypothesizes that they made their contribution as the planets were forming rather than late in planetary development. If water came from millions of comets or small asteroids, the same steady celestial rain would have bombarded Mercury, Venus, Earth, and Mars, so they would all have begun with the same water characteristics, he says. However, the waters of those four planets now have dissimilar profiles, Owen and other geochemists have found.

photo

BEFORE THE DELUGE. Dozens of dry moon-size objects could have been fusing together to form Earth when, by chance, a cosmic water balloon barreled in.

D. Dixon

If, on the other hand, a relatively small number of planetary building blocks brought water into the inner solar system, chance would dictate whether any one of them glommed onto an embryonic planet. A chance encounter—literally an accident in space—could have essentially flooded a planet in one big splash, but according to the luck of the draw, other planets could have been spared. This could explain the current planets' differences in water content and why no existing objects appear to have been in the recipe for Earth.

To carry so much water, the impactor that doused Earth must have come from between Mars and Jupiter, Morbidelli says. Computer models that he and his colleagues described in the Oct. 1, 2001 Icarus show how this might have happened.

The researchers began with the premise that early in the solar system's formation, scores of planetary embryos about the size of Earth's moon were scattered around the sun to a distance four times that between Earth and the sun now (4 AU). The embryos' gravitational interactions with each other and with a growing Jupiter would have caused their orbits to begin crossing.

Some of these bodies would have collided with each other, building into ever-larger embryonic planets. Eventually, the researchers' simulations show, "out of a hundred or more embryos, just a few terrestrial planets form between 0.5 and 2 AU" from the sun, says Morbidelli. Each planet's unique mix of building blocks includes some embryos from outside its final orbit. In some cases, one or more embryos hail from far enough out that they would have been wet.

The weak point in Morbidelli's model is that there's no way to test whether a chance water delivery occurred in the case of Earth, Drake says. The carrier's elemental and isotopic characteristics would have to have been unlike those of any object that researchers have yet found in the solar system. "You can't rule out a [planetary building block] crashing in at 4.5 billion years ago, but it … doesn't seem geochemically plausible," he says.

Only more data, especially more information about the amount and composition of water on Mars, will resolve the mysterious history of the inner solar system's water, says Jonathan I. Lunine, a planetary scientist at the University of Arizona in Tucson. "If Earth got its water locally"—as Drake and Righter suggest—"then Mars [too] should have been swimming in water," Lunine says.

Preliminary data from the Mars Global Surveyor mission suggest that the Red Planet has large deposits of water (SN: 3/9/02, p. 157). Further analysis of Mars could indicate how much water the inner planets received from common sources, such as comets and meteorites. It could also help scientists characterize the sources of the remainder of Earth's original water budget.

Scientists are also counting on data from future comet encounters. Contour, an unmanned NASA probe scheduled for launch in July, will rendezvous with at least two poorly studied comets. It will pass Encke in November 2003 and then Schwassmann-Wachmann 3 in June 2006. Then, NASA may park the probe in a distant orbit to observe any other comets that come by. Data from the close encounters will give scientists better information on the noble gases in comets and could indicate how much cometary material ended up on Earth.

If any comets are found to have Earthlike deuterium-hydrogen ratios, they could add power to the late-veneer theory. Delsemme maintains that the comets responsible for the late veneer formed closer to the sun than the bulk of those left today—and thus had unique isotopic signatures. If he's right, then perhaps our oceans aren't a product of a rare celestial accident after all.

********

References:

Delsemme, A.H. 2001. An argument for the cometary origin of the biosphere. American Scientist 89(September-October):432–442.

Drake, M.J., and K. Righter. 2002. Determining the composition of the Earth. Nature 416(March 7):39–44.

Laufer, D., … and T. Owen. 1999. From the interstellar medium to Earth's oceans via comets—an isotopic study of HDO/H2O. Icarus 140(Aug. 1):446–450.

Morbidelli, A., … J.I. Lunine, et al. 2000. Source regions and timescales for the delivery of water to the Earth. Meteoritics and Planetary Science 35:1309–1320.

Robert, F. 2001. The origin of water on Earth. Science 293(Aug. 10):1056.

Further Readings:

Cowen, R. 2002. Odyssey's First Look: Craft spies signs of ice at the Martian south pole. Science News 161(March 9):149. Available at Science News.

_________. 2002. Martian equator: A watery outpost? Science News 161(March 9):157. Available at Science News.

_________. 2001. A Rocky Bicentennial. Science News 160(July 28):61–63. Available at Science News.

Monastersky, R. 1996. Pursuing life on two frontiers: On Earth. Science News 150(Nov. 9):292. Available at Science News.

Perkins, S. 2002. Space Rocks' Demo Job: Asteroids, not comets, pummeled early Earth. Science News 161(March 9):147–148. Available at Science News.

Sources:

Armand H. Delsemme
2509 Meadowwood Drive
Toledo, OH 43606

Micheal J. Drake
Lunar and Planetary Laboratory
University of Arizona
Tucson, AZ 85721

Sergei Ipatov
Department of Physics and Astronomy
5C3, CSI
George Mason University
Fairfax, VA 22030-4444

Jonathan I. Lunine
Lunar and Planetary Laboratory
University of Arizona
Tucson, AZ 85721

Alessandro Morbidelli
Observatoire de la Côte d'Azur
Boulevard de l'Observatoire
B.P. 4229
F-06304 Nice Cedex 4
France

Tobias C. Owen
Institute for Astronomy
University of Hawaii
2680 Woodlawn Drive
Honolulu, HI 96822

Leonid M. Ozernoy
Department of Physics and Astronomy
5C3, CSI
George Mason University
Fairfax, VA 22030-4444

Kevin Righter
Lunar and Planetary Laboratory
University of Arizona
Tucson, AZ 85721

Francois Robert
Museum-Mineralogie CNRS
61 Rue Buffon
75005 Paris
France


From Science News, Volume 161, No. 12, March 23, 2002, p. 184.