Science Fact: PLATE TECTONICS, GOLDILOCKS, AND THE LATE HEAVY BOMBARDMENT: WHY EARTH ISN'T MARS OR VENUS by Richard A. Lovett
In my part of the world, weather forecasts are the subject of much joking—hard to get right, sometimes even as little as twenty-four hours in advance. But in the billion-year time frame, the Earth's weather is easier to predict. Warm but not too hot, cool but not too cold. Not much chance of freezing the oceans solid or evaporating them away completely.
This didn't have to be the case. Mars and Venus appear to have started out quite Earthlike, but then veered into climate dead-ends: one frozen, the other baking in a runaway greenhouse. Why then is Earth's climate so stable? Even when something tips it out of whack—like the giant ice age of Snowball Earth (in which the ice caps spread nearly to the Equator[1]), our planet recovers.
Partly, of course, this is a side effect of a lucky position in the Solar System. Not only are we in the Sun's habitable zone, where planets won't automatically wind up like Mercury or Pluto, but of the three worlds in this zone, we're the Goldilocks planet. “In very simple terms,” says Victor Baker, a planetary geologist at the University of Arizona's Lunar and Planetary Laboratory, “Venus is too hot, Mars is too cold, and Earth is just right."
But that's just the starting point. In the 4.5 billion years since the Solar System formed, the Sun has brightened by about 25%, but our climate has barely changed. Obviously, something is keeping it stable.
In the old days, scientists and science fiction writers credited the Earth's oversized moon with our planet's habitability. The theory was a bit vague, but was based in part on the observation that the Earth and Moon are a unique pairing.[2] Venus is moonless, and the two moons of Mars, Phobos and Deimos, are tiny. Maybe, the theory went, life-supporting planets, here or in other solar systems, might also need big moons.
In fact, the latest theories indeed suggest that the Moon plays a role—something we'll discuss later in this article. But two other factors are much more important: plate tectonics and water. In combination, these interact with carbon dioxide to form a very effective global thermostat.
This process begins with volcanoes.
Volcanoes spew out more than molten rock. They also produce gas, a major component of which is carbon dioxide. As anyone familiar with the global warming debate knows, carbon dioxide is a “greenhouse gas” that traps infrared radiation that would otherwise escape to space. Glass does the same thing, which is why greenhouses can stay warm without heaters and why the inside of your car gets so hot in the summer.
Greenhouse gases are important to maintaining a habitable Earth. Without them, scientists estimate, our planet would not have been able to rise above the freezing point for much of its early history.[3]
Volcanic gases, however, work in only one direction—warming the planet. A habitable planet can't just let them build up forever; it needs something to remove excess carbon dioxide from the atmosphere. That something is rain.
Rain dissolves carbon dioxide from the air, creating a weak acid (carbonic acid) that falls on exposed rocks, reacting with them in a process called atmosphere weathering. This not only slowly dissolves the rocks, but it creates carbon-containing byproducts (bicarbonates) that then wash out to sea.[4] There, the bicarbonates precipitate out of the water and accumulate on the seabed as carbonate rocks such as limestone.
This gives us two opposing processes. One (volcanoes) injects carbon dioxide into the air. The other (rain) removes it.
What makes this such a good thermostat is that the two processes react differently to changes in temperature. Volcanoes are unaffected. Over the long run, they inject carbon dioxide into the atmosphere at a more-or-less steady rate. Atmospheric weathering, on the other hand is highly dependent on temperature. When the planet is hot, more water evaporates from the oceans. More rain falls, increasing the rate of weathering and reducing the amount of carbon dioxide in the air. When the planet is ice-age cold, the process practically grinds to a halt, allowing volcanic carbon dioxide to build up in the air, rewarming the planet. The result: a tendency for even big climate fluctuations to revert to a happy medium.[5]
* * * *
Recycling the Seabed
So far, though, we've not described a process that can operate on a billion-year timeframe. All we've got is a one-way process in which carbon dioxide goes from volcanoes to the atmosphere to the seabed. Without a way to replenish the carbon dioxide going into the volcanoes, they'd eventually run out. Atmospheric carbon dioxide levels would dip toward zero and the Earth, once warm, would go into the freezer.
That's where plate tectonics enter the picture. I've discussed this in other articles,[6] but it's worth another quick summary here.
In simplest terms, plate tectonics is the cause of “continental drift.” That's because the Earth's crust is composed of plates that slowly move in response to deep currents in the Earth's mantle. Where these currents rise to the surface, plates get pulled apart. Elsewhere, plates collide. The part of this process that's relevant to the global thermostat is “subduction,” in which seabed plates are sometimes shoved beneath continental plates. This pushes the seabed into the Earth's interior, carrying carbonate sediments with it. There, the Earth's heat bakes the carbon dioxide out of the sediments. It mixes with magma, rises to the Earth's surface, and erupts in volcanoes.
We now have a simple but very important cycle: volcanoes, carbon dioxide, rain, atmospheric weathering, carbonate rocks, subduction, and more volcanoes. It's a cycle in which carbon dioxide is not only used over and over again, but the Earth's temperature tends to stay near that happy medium. It's a process that's operated on Earth through most of our planet's history and is likely to continue into the distant future—so long as we continue to have both plate tectonics and a liquid ocean. But it's not operating on Mars and Venus. Why were we lucky, when our neighboring planets weren't?
* * * *
Disappearing Water
In the case of Venus the answer probably lies in the fact that there's a limit to how much solar heat the water/tectonics/carbon-dioxide cycle can accommodate.
Although there are many mysteries about Venus, scientists think it probably started out very much like the Earth, with liquid water, rainfall, and plate tectonics. But it was just enough closer to the Sun to get in trouble.
The problem is that carbon dioxide isn't the only greenhouse gas. Water vapor is, as well. Thus, there are two competing effects when an ocean-bearing planet gets warmer. One is the carbon dioxide/rainfall thermostat. But opposing that is the effect of increasing water vapor. At low-to-moderate temperatures, that's minor. But as the planet warms, the amount of water vapor in the atmosphere rises exponentially. “That makes for a stronger greenhouse, which leads to a hotter surface and more evaporation,” says David Grinspoon, curator of astrobiology at the Denver Museum of Nature & Science. “Once that runaway's going, it's hard to stop."
Eventually, Venus became so hot that essentially all of its water was in the atmosphere as vapor, producing a super-greenhouse that kept the surface too hot for rain to fall. That, in turn, allowed carbon dioxide to build up, making the planet hotter yet.
Meanwhile, water vapor was making its way into the upper atmosphere, where solar radiation could break it into hydrogen and oxygen (a process that has been observed going on today by instruments on the Venus Express spacecraft, currently in orbit around Venus).[7] Hydrogen is too light a gas for terrestrial-sized planets to retain, so it escapes into space, leaving Venus with only traces of its original water. (The oxygen reacts with other materials and is removed from the atmosphere.)
"That means no more thermostat,” says Grinspoon. “You don't have any way to cool off."
That's Venus. Mars is a different story. Scientists once thought that it, like Venus, had also lost its water early on—or that it had never had much to begin with. But from the moment we started flying probes to the Red Planet, that theory started looking doubtful.
To start with, Mars has thousands (and probably millions) of miles of river channels. In a 2008 study, a team led by Brian M. Hynek of the University of Colorado's Laboratory for Atmospheric and Space Physics painstakingly counted all the channels revealed by the latest high-resolution satellite photos. Their tally: 40,005 distinct river valleys.[8] Not only did this quadruple the known number of valleys, but some were seventh-order streams, meaning they were tributaries of tributaries of tributaries ... all the way up though seven tiers of stream branchings. That's important because it indicates that these watersheds were formed by a dispersed source, rather than scattered springs. In other words, rain.
But simply counting valleys wasn't Hynek's main goal. Rather, he asked, “Can we figure out the last time it rained and formed valleys on Mars?"
The answer is a qualified yes. The new satellite images are good enough that it's possible to count the numbers of craters in creek beds—a standard technique for determining the age of planetary landscapes. The more craters, the longer an area has been exposed to meteor bombardment and the older its surface must be. Hynek concluded that some creek beds may have been formed as recently as half a billion years ago, but these look to be of the type created by seepage from springs. The last ones that were clearly precipitation-fed were formed about 2.8 billion years ago.
That's a long time ago, but Mars is 4.5 billion years old. Thus, for more than one-third of its history, Mars had enough water for at least scattered rainfall.
Nor has this water vanished. We've long known that Mars has ice caps, but they're only a fraction of the amount of water needed for an Earthlike planet. More water appears to be hidden in permafrost, some quite close to the surface. In 2008, for example, the Phoenix lander found ice crystals in a trench scratched only a few inches into the surface of the Martian arctic. Additional permafrost is widely believed to lie more deeply buried, elsewhere.
There may even be mountain glaciers buried under thin layers of rock and dust. In several places in the Martian mid-latitudes, roughly between 35 degrees and 55 degrees, either north or south, oddly lobed “debris aprons” spill from crater rims. Photos show them to be rock-covered, but their shapes suggest they were formed by something flowing downhill, something that behaved a lot like glacial ice.
In another study reported in 2008, scientists got a chance to look at two of these debris aprons with ground-penetrating radar carried on NASA's Mars Reconnaissance Orbiter.[9] In a paper in Geophysical Research Letters (J.J. Plaut, A. Safaeinili, J. W. Holt, R. J. Phillips, J. W. Head, R. Seu, N. E. Putzig, and A. Frigeri, “Radar evidence for ice in lobate debris aprons in the mid-northern latitudes of Mars,” 28 January 2009) and presented at the fall 2008 meeting of the American Geophysical Union, Jeffrey J. Plaut of NASA's Jet Propulsion Laboratory and an international team of coworkers found that whatever lies beneath the rocky surface is virtually transparent to radar to a depth of about 500 meters, indicating that it's probably made of 90% pure water ice. The volume of ice isn't huge, but it's close to the surface (within 10 meters, Plaut estimates from the radar signal): yet another sign that a great deal of water still exists on Mars, scattered around the planet in icy deposits.
"Mars had, and still has, lots of water,” Baker says simply.
All of this, plus the copious river channels, suggests that Mars spent the first half-billion years or so of its history wet and fairly warm. “It may well be that Mars, early in its history, had something like [the Earth's] thermostat,” Grinspoon says. And whatever killed the thermostat wasn't a loss of water. Rather, it appears to have been a freezing-up of plate tectonics.
One of the signs of plate tectonics on Earth is the existence of magnetic “stripes” in the seabeds, marking bands of volcanic rock that oozed up along mid-ocean ridges as tectonic forces gradually widened the seabed. These rocks carry the imprint of the Earth's ancient magnetic field at the time in which they solidified, and because the magnetic field varies with time, the rocks’ magnetic pattern also varies, with similar-age rocks having the same pattern. Using this, geophysicists can trace the history of the ocean's spreading.
By 1999, scientists were finding similar magnetic striping in the oldest rocks of Mars, a strong sign that they had been formed in a similar manner, as magma oozed up from below to fill gaps formed by the spreading of tectonic plates.[10]
But if Mars was indeed born with plate tectonics, they appear to have shut down quite early in its history. There are certainly no signs of geologically recent tectonic activity.
To understand what went wrong, we need to look a bit deeper into the planet's interior.
Plate tectonics, as we noted earlier, is driven by currents in the mantle. But these currents are themselves driven by heat escaping from the planet's core, so ultimately it is heat from the core that is responsible for tectonics.
Mars is substantially smaller than the Earth, which means its core is also smaller. When it was young, that wouldn't have made much difference, but with the passage of time, its core would have cooled more quickly than the Earth's, reducing the heat supply to the mantle. Less heat flow means less-vigorous mantle currents. That in turn means less-vigorous tectonics and a crust that steadily thickens as it, in turn, cools. Eventually, it becomes too rigid to move and tectonics grinds to a halt.
Grinspoon compares Earth and Mars to baked potatoes. “Big ones stay hot longer,” he says. “It's the same with planets."
The smaller size also made it harder for Mars to retain a thick atmosphere. “It probably lost more than 99% of its original atmosphere,” Grinspoon says. “You simply can't support a strong greenhouse effect when you have that thin an atmosphere."
* * * *
Death of A Dynamo
In part, the atmosphere loss would simply have been caused by gas escaping the weak gravity. But the core may also have played a role by switching off the Martian magnetic field early in the planet's history.
Planetary magnetic fields are created by convection currents in a planet's molten-iron core. Electrically charged particles in these currents act as a dynamo, inducing a magnetic field from their motion.[11] We know that Mars once had a magnetic field, thanks to the magnetic stripes seen in the early rocks—something that can only be created if they formed in the presence of a magnetic field. But later rocks aren't magnetized, and by comparing them, we can estimate that the dynamo shut off when the planet was only a few hundred million years old.
The end of the dynamo is important because a planet's magnetic field does more than simply imprint its signature on solidifying rocks. It also shields a planet from energetic charged particles in the solar wind—particles, which, unblocked, can speed the dissociation of water in the planet's atmosphere.
Mars isn't Venus. We know that it retained most of its water. But Mars was never at risk of overheating; only a small fraction of its water was ever in the atmosphere at any given time. When the dynamo stopped, it would only have been this vapor that would have been exposed to erosion by the solar wind. Surface water, whether in oceans or ice, wouldn't be directly affected.
But while the atmospheric water vapor would only have been a small fraction of Mars's total water supply, it was the part that was contributing to greenhouse warming. Thus, as its core cooled, Mars was hit by a double whammy. At ground level, plate tectonics shut down, reducing the supply of fresh carbon dioxide. Higher up, water vapor was being lost to space. Eventually, everything froze, giving us the planet we know today.
Of course, few things in comparative planetology are ever that straightforward. Mars's small size undoubtedly played a role in the demise of its tectonics. But other factors might have contributed. Maybe the Martian core is different from the Earth's. If Mars has less iron than the Earth, for example, its core would be under-sized even for the size of the planet, and it would have had an even harder time sustaining plate tectonics.
Another theory is that the core may have been struck a fatal blow during the late heavy bombardment.
Planetary scientists believe that the Solar System formed in several stages. In the first, the Sun's protoplanetary disk condensed into a multitude of balls of primordial planet-stuff: fluffy dust-balls, perhaps the size of tennis balls. These then quickly coalesced via a series of collisions into bigger objects, some of which became the planets we know today. The first collisions would have been among small objects, but as the growing planets swept up all the small stuff, they would have wound up colliding with ever-larger objects that had themselves been growing for some time. Crater counts on Mars and the Moon (among other objects) support this theory: there came a time, near the end of the planet-forming epoch, when collisions were big.
On the Earth, one of these, with a Mars-sized object, may have formed the Moon.[12] On Mars, a collision with a 1,000-mile to 1,800-mile object may have created the Borealis basin: a northern lowland, covering roughly 40% of the planet.[13]
Overall, says James Robertsof the University of California, Santa Cruz, we can count about twenty really large impact basins on the Martian surface. It's possible, he added at the fall 2008 meeting of the American Geophysical Union, to assign dates to these by counting the number of smaller impact craters on their floors. (A lot of them are 4.1 to 4.2 billion years old, presumably representing the heaviest part of the late heavy bombardment.) We can also use satellite measures to determine the degree to which these basin floors are magnetized. When that's done, Roberts said, a striking pattern emerges. The oldest craters have magnetized floors; the later ones are unmagnetized. Sometime during the late heavy bombardment, the Martian magnetic field seems to have shut off—rather abruptly, it would appear.
Is that a coincidence, or did asteroid impacts kill the dynamo?
Possibly the latter, says Roberts. The mechanism would have been the heat of impact, which would have been at the surface, not at the core. This would have greatly reduced the temperature gradient between the core and the surface, possibly even temporarily reversing the direction of heat flow, with heat going downward rather than upward.
According to his computer models, Roberts said, a weak dynamo turns out to be easy to shut down forever. “You can kill it with a very small drop in heat flow,” he said, “but it takes a lot to restart it."[14]
The same aberration in heat flow might also have affected mantle currents for long enough that they, too, did not recover. Thus, a giant impact in the late heavy bombardment (or a succession of impacts) may have been all it took to shut down the Martian magnetic field, setting in motion the processes that eroded the Martian atmosphere,killed plate tectonics, and produced the planet we see today.
* * * *
Wayward Spins
So far, we've found that the Earth was fortunate in two ways: it wasn't too close to the Sun, and it was large enough for plate tectonics and a core dynamo to continue for billions of years.
Three other factors also helped. One is simply the fact that the Earth has life.
Life has radically changed the Earth. “If you had done an environmental impact statement four billion years ago and said, ‘Should we let life start?’ you would have said ‘No,’ because it's going to completely screw up the environment,” jokes James Head III, a planetary scientist from Brown University. “It [life] has radically changed [the environment]. For the better, we might say, but who knows what was asking the question then!"
One of the things life has done is infuse the atmosphere with oxygen, something that wouldn't have happened without photosynthetic plants and bacteria. But that's not the only effect. “A lot of [marine] organisms secrete calcium carbonate shells, and they get deposited in the ocean,” says Head. “That takes a huge amount of carbon dioxide out of the atmosphere."
Another factor that has helped is simply the continued existence of large quantities of water. Partly that's just because the oceans are a huge heat sink, capable of damping short-term variations in climate. But some of that water is dragged beneath the surface in subduction zones via waterlogged sediments, where it helps lubricate the process of plate tectonics.[15] “It's not well known how it acts,” says Michael Mischna, a planetary scientist at NASA's Jet Propulsion Laboratory, “[But] it allows plates to slip and slide against each other. When you lose that, it locks everything up, like an engine seizing."
The third factor is the Moon. The old theory was that its gravity somehow stripped off excess atmosphere, keeping us from getting Venus-style hot. The new theory is that the Moon helps keep ice ages from becoming too severe. The best way to see how this works is to look at Mars.
Mars, Head and other scientists are finding, has long undergone substantial wobbles in its spin axis—wobbles that have wreaked havoc on its seasons.[16] Working backward in time, he says, it's possible to calculate the effect of perturbations from other Solar System bodies on the tilt of Mars's axis. “You can see it's whipping like a whiplash,” he says.
In the past few million years, he's found, it's ranged from very little tilt to an angle of 35 to 40 degrees. In fact, he says, there's a reasonable chance Mars could practically flip over on its side, tilting by 70 or 80 degrees.[17]
The Earth also sees variations in its tilt, but only by a few degrees. Even these are large enough to launch (or end) ice ages. The difference, Head and others say, is that the Earth's spin is stabilized by the Moon. “Mars has just these two little space potatoes, Phobos and Deimos, which don't really do much,” Head says.
Mischna adds: “Without something like the Moon, essentially locking the Earth into place, the axial tilt tends to wobble back and forth. That's what we see on Mars: a very large wobble. We have only very subtle changes. That keeps the planet pretty much stable relative to Mars."[18]
* * * *
So, what of the future?
In the long run, the Sun is continuing to heat up and eventually Earth's climate will go the way of Venus's.
"Venus shows us what happens to an Earthlike planet when the input of solar radiation gets over a certain threshold,” Grinspoon says.
But there's no hurry building a spacefaring ark: the Sun probably won't get hot enough to boil off the oceans for a billion or more years.
As for our own effects on climate, the good news is that in the million-year long run we're more likely to ruin ourselves than the Earth's climate as a whole. But that doesn't mean it's impossible to do both. “Both Venus and Mars seem to have gone from habitable to uninhabitable,” Grinspoon says. “That should give one pause about messing with one's planetary climate without a complete understanding of the limits of that stability."
Copyright ? 2009 Richard A. Lovett
* * * *
1 See R. A. Lovett, “From Fimbulwinter to Dante's Hell: The Strange Saga of Snowball Earth,” Analog, Jan/Feb 2006.
2 Technically, the Earth and Moon aren't quite unique. Pluto’ moon Charon is also large, and other Kuiper Belt objects come in pairs. But none of them are in the habitable zone.
3 For more on this, see an un-bylined article in the online edition of Astrobiology, February 7, 2007.
4 Environmental concerns about acid rain involve another acid, sulfuric acid, produced, in part, by the burning of sulfur-containing fuels.
5 Ideally, we could rely on this process to abate our current climate concerns. Unfortunately, it's slow by human standards, requiring, by most models, thousands of years to remove the extra carbon dioxide we're injecting into the air by burning fossil fuels.
6 See, for example, “Messengers from the Earth's Core? The Great Plume Dabate Heats Up,” Analog, July/August 2006.
7 Press release from the European Space Agency, dated 18 December 2008. Such press releases can be found at www.eurekalert.org.
8 This study was reported at the fall 2008 meeting of the American Geophysical Union. Abstracts for this, and other studies from that meeting, can be found at www.agu.org.
9 The same method has been used to map layers in the Martian ice caps, all the way down to bedrock. See www.nasa.gov/missionpages/MRO/multimedia/phillips-20080515.html.
10 See Richard A. Kerr, “Signs of Plate Tectonics on an Infant Mars,” Science 30 April 1999, pp. 719-722.
11 For a more detailed explanation, see: sunearth.gsfc.nasa.gov/sunearthday/2004/vtdynamoplaentary2004.htm. [Sic: the typo “plaentary” is in the URL.]
12 For more, see www.space.com/scienceastronomy/planetearth/moonwhackside000901.html.
13 See: R. A. Lovett, “Texas-Sized Asteroid Slammed Early Mars,” National Geographic News (online), June 25, 2008.
14 In Roberts’ model, as little as a 1% drop in heat flow from the core can be enough to shut down the dynamo. Restarting it would require the heat flow to rebound by a full 25%.
15 In 2007, for example, seismologists calculated that an enormous mass of squishy rock beneath China contains an amount of water comparable to the Arctic Ocean. See, R. A. Lovett, “Huge Underground ‘Ocean’ Found Beneath Asia,” National Geographic News (online), February 27, 2007.
16 Traces of these cycles appear in several ways. One is in the carving of gullies by what appears to be melting snow, which, Head and coworkers believe, appears to have fallen repeatedly in certain highlands over the past few million years. Only massive climate cycles—the Martian equivalent of ice ages—would allow snow to melt from the ice caps, fall in the highlands, then melt again. Other, more ancient, evidence appears in “layered deposits” in a region called Arabia Terra. In a presentation at the 2008 Lunar and Planetary Science Conference in League City, Texas, Kevin Lewis of CalTech used high-resolution orbital photos to examine these sediments in three dimensions, at one-meter resolution. His team found a regularly repeating succession of layers, indicating that Mars was probably undergoing regular climate cycles even very early in its history, when these sediments were being laid down.
17 See www.planetary.brown.edu/htmlpages/publications.htm.
18 Venus, on the other hand, has no moons but no substantial wobbles, possibly because of its much slower spin (Venus rotates once every 243 Earth days.) Thus, there are spin states, Grinspoon says, that are stable without the presence of a large moon. But in the case of Venus, the price is very long days. Also Venus has no significant axial tilt, which means no seasons.