Ebb and Flow, the twin Gravity Recovery And Interior Laboratory (GRAIL) spacecraft launched by NASA in September 2011, have revealed unexpected details about the Moon's interior.

Update: Having completed their extended missions, the GRAIL probes were intentionally crashed into the Moon on December 17th. They hit the lunar surface 30 seconds apart, at 5:28:51 p.m. EST and 5:29:21 p.m. EST, striking the southern face of a tall, unnamed mountain at 75° 37′ north, 26° 38′ east (northwest of the crater Goldschmidt). NASA also announced that the collision sites would be named to honor the late astronaut Sally Ride.
Only a moment's gaze at the Moon through a small telescope tells you pretty much everything you need to know about its history. The bleak lunar surface was heavily battered by collisions large and small soon after it formed. Some crash sites fractured the crust so deeply that torrents of magma gushed from the interior, creating the Moon's distinctive dark lava "seas." And nothing much else has happened there in the past 3 billion years. Over the past 50 years, a few dozen robotic spacecraft (and of course the rock collecting of a dozen Apollo astronauts) have refined many details of lunar history. For example, scientists know the near and far sides look very different, and they now have a much better handle on how the Moon formed (most likely involving the titanic impact of a planet-size body with early Earth). Yet our basic understanding lunar surface geology has remained unchanged for decades.
GRAIL's gravity maps

GRAIL's "free-air" gravity map (left) shows deviations caused by both the Moon's bumpy surface and its lumpy interior. Compare that to the Bouguer gravity map (right), which removes effects of topography to reveal density variations underneath the surface (such as mascons underlying large impact basins). These views show the lunar far side, centered on 120° west.

NASA / GSFC / Science Visualization Studio
Against this backdrop, yesterday a team of planetary scientists unveiled new findings about the Moon that not only were unexpected but will also alter some long-held notions about our satellite's early history. Apparently the lunar exterior has been cracked and pulverized so violently and completely that it's a total jumble of rubble at least to a few miles down.

Just as a rock pile on Earth is full of gaps and spaces, so too is the upper lunar crust. According to investigator Mark Wieczorek (University of Paris), on average this pulverized outer layer has a porosity of 12% — though he excluded the maria, because their high density would bias the measurements. The porosity is even greater (up to 20%) around relatively fresh mega-impacts like the far side's Orientale and Moscoviense basins.

Also, although the lunar crust was presumed to be 40 miles (60 km) thick in post-Apollo days, and somewhat less than that based on more recent spacecraft results, it now seems that nowhere is the Moon's rigid outer skin thicker than about 27 miles (43 km) — and quite a bit less in many spots. Moreover, its density is much lower than thought, averaging only 2.55 g/cm3. This thinner skin implies that the Moon's bulk allotment of aluminum (an element common in low-density silicate rocks) must be very comparable to Earth's, notes Wieczorek — yet another line of evidence reinforcing a "big splat" lunar origin.

Thickness of the Moon's crust

GRAIL data of the Moon's near side show that the lunar crust is not as thick as once thought — and it's especially thin where major impacts have blasted out huge basins.

NASA / GSFC / Science Visualization Studio

So what's the source of these revelations? They are courtesy of identical spacecraft named Ebb and Flow, the business end of NASA's Gravity Recovery And Interior Laboratory mission. GRAIL's objective is to map the Moon's gravity field. Those of you "of a certain age" might recall that this has been done before, beginning with the first orbiters in the 1960s. Back then NASA flight controllers noticed that their spacecraft were periodically accelerating and decelerating very slightly while orbiting the Moon. Apparently specific lunar features, most notably some of the lava-topped maria, harbored concentrations of mass (mascons) that weren't obvious to the telescopic eye. Since then other orbiters, notably NASA's Lunar Prospector and Japan's Kaguya, have produced ever-better gravity maps.

But GRAIL's new gravity maps are an eye-popping 100 to 10,000 times more detailed than previous efforts. During its prime mapping phase, from March 1st to May 30th, Ebb and Flow were never separated by more than 135 miles (218 km) as they chased each around around the Moon at a mean altitude of just 35 miles (55 miles).

GRAIL spaceraft around the Moon

Using a precision formation-flying technique, the twin GRAIL spacecraft measure irregularities in the Moon's gravity — key to revealing the size of the Moon's core and other longstanding questions about the lunar interior.


Here's how it works: When the lead craft is drawn toward a mass down below — be that a high-standing crater rim or a buried mascon — it accelerates slightly farther ahead of its sibling and vice versa. The key to GRAIL's success is that the craft kept track of each other's whereabouts five times per second — and they do so very, very accurately, detecting velocity changes as small as 50 nanometers per second. "That is one twenty-thousandth the velocity that a snail moves," comments Maria Zuber, the MIT geophysicist who serves as the mission's principal investigator.

In fact, the just-released gravity maps are so good, she notes, that they resolve the tiny gravitational effects from the high rims and low floors of every lunar crater at least 20 miles (30 km) across and a great many smaller ones.

The new results appear online in December 5th's Science Express. Those three articles are accessible by subscription only, but you can freely view and download a series of terrific global views and animations of GRAIL results prepared by the Science Visualization Studio at NASA's Goddard Space Flight Center.

At NASA's briefing, investigator Jeff Andrews-Hanna (Colorado School of Mines) described the discovery of narrow high-density structures up to 200 miles long that crisscross the surface. These appear to be dikes, thin veins of solidified magma buried globally beneath the lunar surface. They appear to be very ancient, because they're frequently interrupted by craters. "There'll be a lot of work to understand what these dikes mean for lunar evolution," notes Andrews-Hanna.

In order for the dikes to form, the young lunar crust must have become fractured when the interior expanded and pushed outward. All told, the Moon's diameter appears to have increased by up 6 miles (10 km). By implication, the lunar interior must have have started out cooler than the exterior; then the interior got hotter and expanded — but only after the crust had solidified. (Theorists think this outside-in heating scenario is precisely what would have occurred if the Moon built itself up from collisional debris.)

The GRAIL mission isn't over, at least not quite yet. A few months ago, flight controllers cut Ebb and Flow's altitude by half, to an average of just 14 miles (23 km), and that should yield another fourfold improvement in the gravity map's resolution.

But there's a penalty for skimming so close to the crater-tops: "We need to do three maneuvers a week just to avoid crashing on the Moon," Zuber cautions. In just a few weeks, the craft will be allowed to crash into the Moon — most likely targeted and timed so that the eagle-eyed Lunar Reconnaissance Orbiter can watch.


Image of Tony Flanders

Tony Flanders

December 7, 2012 at 7:35 am

Kelly, does this rock jumble pervade the entire surface, or is it just atop the lunar highlands? The maria certainly look smooth, not jumbled, through a telescope.

You must be logged in to post a comment.

Image of Peter


December 7, 2012 at 8:23 am

Is it just me? I cannot identify any familiar lunar features in their images. They look like they could be of Mercury or Callisto.

You must be logged in to post a comment.

Image of Kelly Beatty

Kelly Beatty

December 7, 2012 at 8:32 am

Tony: you raise a good point. the maria were specifically excluded in the porosity estimates. Peter: the top two maps are of the lunar far side, so it's not surprising that nothing's recognizable. but the map of crustal thickness shows some familiar features, which are labeled.

You must be logged in to post a comment.

Image of Bruce


December 8, 2012 at 11:22 am

Since so much of the Moon’s crust is found to be fractured and porous is it not possible or perhaps even likely that there are caves on the Moon? Some underwater cave divers might even wish to ad orders of magnitude to the danger and expense of their hobby 🙂

You must be logged in to post a comment.

You must be logged in to post a comment.