I've been teaching a class of high-school seniors about how the Moon's spin rate and orbital period became locked together, in what astronomers call synchronous rotation. Thanks to a combination of slight irregularities in the lunar interior and Earth's ocean tides, one hemisphere of the Moon must continually face toward Earth and the other must always be turned away from us. Sure, the Moon nods a bit this way or that, due to libration, but basically we're stuck with looking at its near side permanently.

However, a new study, about to be published in the journal Icarus, probes whether there once was no "Man in the Moon" to gaze upon.

Moon's far side

A view of the Moon unseen from Earth but captured in 1972 by the crew of Apollo 16. Familiar Mare Crisium sits along the top limb; below it are Mare Smythii (left) and Mare Marginis. Below them is the heavily cratered lunar far side.

NASA

Mark Wieczorek and Mathieu Le Feuvre, researchers at the Institute of Earth Physics of Paris, find that the Moon's eastern hemisphere has a statistical excess of big and very old impact basins. Yet the Moon's current orientation argues for just the opposite: its western side, which faces forward as the lunar sphere zips around Earth, should have the lion's share of basins. If the Moon has been synchronously locked this way for its entire history, Wieczorek and Le Feuvre calculate there's only a 0.3% likelihood that this mismatch could have occurred by chance.

So they explore how big a push it would have taken to knock the Moon out of lock and flip it front to back. A big one, it turns out: an object at least 30 miles (50 km) across would have to strike a glancing blow fast enough to gouge out a basin a couple hundred miles (350 to 500 km) in diameter. Of the known lunar basins, only six could have done the trick, and the best candidate seems to be Smythii. It's large enough, 350 miles (570 km) across, and it's on the lunar equator.

The putative strike must have taken place at least 3.8 billion years ago, when the Moon was roughly only half as far away as it is now. The strike wouldn't have changed the Moon's spin axis much, but it would instantly alter the spin rate. Primordial Earthlings would have seen the lunar face swing side to side like a pendulum. If this forced libration exceeded 90°, the Moon would have briefly rolled all the way around.

After about a year tidal torques would have regained control and stopped this merry-go-round. But, as Wieczorek and Le Feuvre note, "It would be only a matter of chance as to whether the same face of the Moon would be directed towards the Earth as before the impact."

The French researchers aren't the first to examine the idea of a lunar flip-flop. Impact specialist H. Jay Melosh (University of Arizona) looked into it more than 30 years ago.

The problem then, as now, is that lunar geologists aren't sure they have a complete census of the most ancient impact basins. Many might lie buried under the vast lava plains that formed the maria; dozens more might be hidden under rubble from more recent strikes. That ambiguity should be resolved when scientists complete the improved topographic maps now being assembled from the Kaguya, Chang'e 1, and Chandrayaan 1 orbiters — fortified by NASA's Lunar Reconnaissance orbiter later this year.

Comments


Image of Rod

Rod

February 11, 2009 at 5:47 pm

Intriguing report and if the Moon orbited about half its present distance some 3.8 billion years ago, the lunar month could be <= 10 days at a distance of about 30 earth radii. Finding fossil record evidence to support this short lunar month and close orbit may be difficult. For an alternative view, see Samec, Ronald G., On the origin of lunar maria, JOC 22(3):101, 2008.

You must be logged in to post a comment.

Image of Allan Holmgren

Allan Holmgren

February 13, 2009 at 8:27 pm

A massive impact like that described to cause a lunar flip could have left a residual ringing which might be detectable.

You must be logged in to post a comment.

Image of FER

FER

February 14, 2009 at 5:44 am

Accompanying this article, you've included an image of the Moon taken from Apollo 16. It appears to show a lunar "far side" that has much mroe dramatic relief than the near side. Here's a comparison of that image with a comparable image of the near side:
http://www.fer3.com/compare2.gif
The difference seems impressive!

In fact, this is a bit of an illusion. Look at the size of Mare Smythii in the image and compare it with a lunar globe or an equal-area map of the Moon. You will find that it is much too large. This photo was in fact taken from a rather low altitude above the trailing side of the Moon shortly after Apollo 13 left lunar orbit. The Moon at the time would have filled the windows of the Apollo spacecraft with an angular diameter greater than 75 degrees. Here's a diagram illustrating Apollo 16's viewing location when the photo was taken: http://www.fer3.com/compare2-geometry.gif

The image is usually printed (or displayed digitally) so that it appears much smaller. In effect, this is a "fisheye lense" effect. Local topography of some medium-sized craters and a few mountains to the east of Mare Smythii have been greatly exaggerated in the image leading to the misinterpretation of it as dramatic "far side" topography. This area is no more dramatic than the highlands on the near side. Note that this is not an intentional trick. It's just an artifact of reproducing a wide-angle photo as if it is a common narrow-angle photo. There are indeed some areas with much greater relief on the lunar far side, but this ain't it, and it wouldn't look like this except from a low-flying spacecraft.

You must be logged in to post a comment.

Image of Testing

Testing

February 14, 2009 at 5:46 am

This is a test.

Is there any way to preserve paragraph breaks in messages?

You must be logged in to post a comment.

Image of A. Szautner

A. Szautner

February 15, 2009 at 3:18 am

This kind of thing just amazes me. The Moon - like any other 3-dimensional object - can (in principle) rotate along any of those axes. WHY doers it necessarily have to be an earth-centered rotation, as gabbed up here? WHY does any rotation automatically have to show the far side of the Moon?

Given that the assumption that impacts will all arrive from random directions ALSO EQUALLY provide for the possibility that the Moon can also rotate (more or less) about the earth-moon line axis, can't it also have "rolled over" like that?

It seems that even researchers have so habitually come to visualize the Solar System (the "ecliptic") and any orbital plane of a planetary satellite system as a "horizontal surface" which is automatically viewed from the north which is automatically presumed to be "above", which forces people to "look down" upon that which they seek to understand. It stunts their growth. Part 1 of 2.

You must be logged in to post a comment.

Image of A. Szautner

A. Szautner

February 15, 2009 at 3:23 am

The Moon, it is well known, has been quite asymmetrical in its mass distribution for a very long time. So long, in fact, that the farside thick crust (relatively low density) and the nearside thin crust associated with the preponderance of maria (relatively high density) will have - through a fairly strong stabilizing tidal force - automatically counteracted any impulse for the Moon to "turn around" over most of its history. But what it CAN'T have done anywhere near as well is offered stabilisation in the earth-moon exis line of rotation: the Moon COULD MUCH more easily have turned "upside-down" - so to speak - than turned "about face": the researchers want AN opposite side to get more impacts. FINE! What makes them insist that such a cirumstance requires the far-side to face the Earth??? Part 2 of 2.

You must be logged in to post a comment.

Image of Fred from Laurel, Md

Fred from Laurel, Md

February 15, 2009 at 5:09 pm

I fail to see that there is any loss in understanding from recognizing that the ecliptic plane defines a sort of 'natural' system of coordinates for doing the vector calulations involved here. Too, it must be recognized that the sidereal rotation (angular momentum) of the Moon lies (very nearly) in that plane. The gyroscopic effect must be taken into account here, because it will distinguish flipping forces that line up with the Moon's existing rotation axis from those that don't. Say a collisional impulse is applied to the Moon at its north pole, and comes from the lunar east side. This will apply a torque-impulse pseudovector directed Earth-to-Moon, and will cause it to tilt, not rightward about the Earth-Moon axis, but rather, tilt the Moon's north pole away from Earth and its south pole toward us. One way to get the Moon to turn, rotation-axis and all, rightward (as seen from Earth) about the Earth-Moon axis, would be to hit it at its north pole directly from behind (or at the south pole from directly in front). But the gyroscopic effect does indeed impart some resistance to rotation of the Moon about the Earth-Moon axis, and so it might be impossible for any single collision to cause the Moon to flip 180 about the Earth-Moon line without completely shattering it. If that's the case, then the only way to get a collision to swap the Moon's east and west sides would be to apply just enough torque about the same axis as the Moon's rotation to make it 'wobble' side-to-side enough that tidal-locking forces would be able to finish the flip.

You must be logged in to post a comment.

Image of Fred from Laurel, Md

Fred from Laurel, Md

February 15, 2009 at 7:21 pm

Kelly, a question. Something isn't making sense to me. When you refer to the "Moon's eastern (or western) hemisphere," do you mean celestial (the old way of assigning these directions, projecting Earth directions out into the sky) or selenographic (the new way, intrinsic to the Moon surface, same as we do on Earth)? The hemisphere that leads in the Moon's orbital motion is on the left (celestial west; selenographic east), looking at the Moon with north up. But it seems to me that the Moon's biggest basins ARE on that side. There's Oceanus Procellarum, the largest basin on the entire Moon (but maybe this isn't an impact basin?), Mare Imbrium, the largest circular basin on the Moon, the Orientale basin, Maria Humorum and Nubium, Sinus Iridum, and Korolev and Baily basins. These together cover considerably more of the surface than the biggies on the other side--Maria Serenetatis, Tranquilitatis, Fecunditatis, Nectaris, Crisium, Smythii, Marginis, and Moscoviense. But I may be looking at it wrong. I guess I'm not sure whether some of these count as impact basins. In any case, there's actually a much bigger difference between the front and back sides of the Moon, as regards big impact basins.

You must be logged in to post a comment.

Image of P Wilson

P Wilson

February 16, 2009 at 7:38 am

“Finding fossil record evidence to support this short lunar month and close orbit may be difficult.” According to Linda Elkins-Tanton, in her book The Earth & Moon, there is not only geologic evidence for a short lunar month, but geologists have shown the expansion of the Moon’s orbit is accelerating, like the universe! “Why does it necessarily have to be an earth-centered rotation?” Current theories have the Moon being created in titanic glancing-blow with a larger body that merged with the Earth, but sent a cloud of debris into orbit. This ring then assembled itself into a single clump. If you picture this ring of debris orbiting the Earth, it is easy to imagine that that ring is going to have very little net-rotation, except with respect to its orbit about the Earth. Luna may have been tidally locked with Earth at birth.

You must be logged in to post a comment.

Image of Rod

Rod

February 16, 2009 at 8:24 am

P Wilson. It has been known for some 40 years or more research that the current tidal dissipation parameter for the Earth-Moon system resuls in the Moon at the Roche limit of about 3 Earth radii, 1.5 billion years ago, not 3.8 billion years ago (supported by 3,000 years of solar eclipse measurements from Assyria tablets too). Claims of finding fossil record evidence for a very low tidal dissipation rate 3.8 billion years ago, rests upon what rocks and where? Plate tectonics erases this evidence if it existed. It would be interesting to see who established fossil record evidence for a lunar month < 10 days as would be required if the Moon was only 30 Earth radii distance. This entire issue concerns how long the Moon has been in orbit so it is no small problem.

You must be logged in to post a comment.

Image of Fred from Laurel, Md

Fred from Laurel, Md

February 16, 2009 at 8:58 pm

Uh, I think I had my directions bass-ackward there. Doh! The orbit-leading hemishpere is on the left (with north up), true, but that's celestial EAST = seleno-WEST. And this IS the side with Procellarum, Imbrium, etc. I just had the east-west labels reversed. So from your article I see that you are using the new (selenographic) directions.

My question still stands, then. Could it be that the study you site was concentrating on basins in the 50 - 150 mile range of diameters, or the like? The Apollo 16 photo you include shows the orbital-trailing side (seleno-east).

You must be logged in to post a comment.

You must be logged in to post a comment.