A watery double-planet: Luna and Terra

Science Magazine recently published a paper that reports that minute quantities of water contained in lunar volcanic glass appear to be identical in isotopic composition to terrestrial water.  According to subsequent press reports, this finding revolutionizes our understanding of the origin of Earth and Moon.  But does it?

Water is a simple molecule, made up of two hydrogen atoms and one oxygen atom.  However, these atoms are not all made the same – they always contain the same number of protons and electrons but the number of neutrons they contain varies.  In particular, some naturally occurring hydrogen contains an extra neutron and hence has twice the mass of normal hydrogen.  This “heavy hydrogen” (called deuterium, for its atomic weight of two) is much less abundant than its lighter version.  Planetary scientists use the amounts of deuterium, relative to normal hydrogen, as a measure of the provenance of the material, i.e., where it formed relative to the Sun. 

Ultimately, substances that have identical deuterium/hydrogen ratios are presumed to have come from the same source.  We have reason to believe this ratio increases systematically outward from the Sun, depending upon where in the early “solar nebula” the material condensed and its subsequent geological processing.  Oxygen (the other element in water) also has an isotopic variation; normal oxygen has 16 protons in its nucleus, but the other isotopes of oxygen can have an additional neutron or two.  As with hydrogen, the variation in the ratios of normal to “heavy” oxygen is thought to be indicative of where the material comes from.

Of course nothing is ever quite so simple and straightforward.  Subsequent processing, such as interaction with cosmic rays, can sometimes alter the composition of samples but if these effects can be accounted for and eliminated, isotopic composition can be used as a tool to map the ultimate sources of Solar System debris.  This has been done with many different elements and compounds, but oxygen and hydrogen are very volatile and thus, sensitive indicators of the thermal environment in which they formed. 

When the isotopic composition of an element like oxygen is plotted for the various groups of Solar System materials – meteorites, lunar, martian and terrestrial samples – they all form distinct groups, indicating that the source reservoirs of these materials formed in different locations of the nebula.  The most primitive type of meteorite – carbonaceous chondrite – appears to have formed at the farthest distance from the Sun.  These rocks are thought to have originated within once icy bodies, the cores of objects known as comets.  Comets form in the outer Solar System where low temperature substances are abundant and are occasionally perturbed by gravity to enter the inner Solar System, i.e., inside the orbit of Jupiter.  Once there, they are heated by the Sun and their most volatile components are sublimed away; after multiple passes through the inner planet zone, only a small fraction of this primitive material remains.

The new findings indicate that the isotopic composition of the hydrogen in water in the mantle (deep interior) of the Moon is nearly identical to that in the water of Earth’s mantle, and both appear to have come from carbonaceous chondrite (most primitive) meteorites.  When compared to a variety of data from other Solar System objects (including the giant planets, icy outer planet satellites and meteorite groups) the Earth-Moon system is compositionally distinct and identical, indicating that, whatever our origins, the description of Earth and Moon as a double-planet is even more appropriate than we had thought.

What does this mean for lunar origin and what does it say about the water at the Moon’s poles?  The bulk composition of the Moon has long been recognized as a key constraint on models of lunar origin.  A basic question is whether the Moon is made of the same material as the Earth or not.  The new results indicate that it is and as such, is another contributory piece of evidence that the materials of the Earth and Moon were brewed in the same pot.  Interestingly, this pot of material is distinct from virtually every other Solar System object (as near as we can tell based on limited information from the other planets).  Whatever process formed the Moon, it involved objects that were created more or less in this neighborhood of the Solar System.  The new results also suggest that both Earth and Moon had a significant component of water early in its history.  Earlier studies had suggested that the terrestrial hydrosphere was a late addition, a veneer of cometary debris from deep space that was added to the Earth late in its history.  We now know that this water was incorporated into the Earth very early, possibly from the beginning of accretion.  The Moon shares this trait – and the same source of water.

So is the giant impact model of lunar origin still viable?  The existence of water in the lunar interior is not a prediction of the giant impact model but as has happened previously, the model will probably be modified to accommodate the new findings.  We have a tendency to imagine (and desire) simple systems in chemical and thermal equilibrium, in which materials and energy behave in a straightforward, predictable manner.  But this event (if it occurred) was a singular one, possibly involving complex, chaotic behavior.  Thus, some of the difficulties created by the new data will probably be explained away.  A hypothesis elastic enough to be stretched to fit any new discordant observation isn’t particularly useful and certainly isn’t scientific.

How does this affect our thinking about the water ice trapped at the Moon’s poles?  As we continue to find that the interior of the early Moon was more water rich than previously thought, we must add lunar water to the long list of possible sources for polar-trapped water.  (As a reminder, the previous idea was that polar water was derived from external sources – the Sun via the solar wind hydrogen, water-bearing meteorites and comets).  Could at least some of the water at the poles be of lunar origin?  One problem that we still don’t understand is the geological age of the polar cold traps – they exist because the spin axis of the Moon is normal to the ecliptic plane.  How long has the Moon been in this orientation?  We suspect that the Moon has been stable for at least the last 2 billion years but water is being found in volcanic glass over 3 billion years old and thus, released before the current polar cold traps existed.  So at least for now, it seems that the Moon’s own water is an unlikely contributor to the ice at its poles.  But that story could change too.

The Moon’s surprisingly complex and interesting history continues to confound the experts.  We may have already “been there” but we still don’t fully understand the Moon’s story and true potential.

Shackleton crater, Moon. Clockwise from top left: topography from laser altimetry, image from SMART-1 mission, lighting map (brighter is longer periods of illumination) from the LRO Camera, Mini-RF CPR image draped over shaded relief map. The crater is about 20 km in diameter.

Though unremarkable in appearance compared to the roughly 4,000 craters on the Moon in its size range, the 20 km diameter crater Shackleton has been the source of relentless scientific controversy for the past 20 years.  Shackleton is located at the south pole of the Moon; indeed, its near side rim is the precise location of the geographic pole itself.   Its location makes observation by Earth-based telescopes difficult and it was not well photographed by the Lunar Orbiter series (our principal source of lunar images) of the 1960s.  That all changed in 1994 with the flight of the joint DoD-NASA mission to the Moon, Clementine.

Clementine carried cameras that globally imaged the Moon in eleven visible and near-infrared wavelengths.  In addition, it mapped the surface and lighting of the poles of the Moon at uniform resolution over the course of almost three lunar days (74 Earth days).  When the Science Team first saw the south polar mosaic, the extent of darkness in the map was striking.  Because the Moon’s spin axis is close to perpendicular to the ecliptic plane, the Sun is always at the horizon at the lunar poles.  Instead of rising and setting, the Sun circles around the poles at or near the horizon.  Because of this grazing incidence, an area in a topographic depression may be in permanent shadow.  And so it appeared for Shackleton crater in the Clementine data, setting off bells in the heads of the Science Team.

A key controversy of the post-Apollo era was whether the lunar poles might contain water or not.  Although the Apollo samples had been studied and found to be “bone-dry,” we had not been to the poles on any Apollo mission.  We knew that any shadowed areas had to be extremely cold as well as permanently dark.  As water-bearing debris in the form of asteroids and comets constantly strike the Moon, it was thought that some of that water might get into a polar “cold trap” and would be kept there (essentially) forever – billions of years of impacting cosmic “debris” can add up.

Clementine was not configured to measure the presence of water, but a cleverly improvised experiment used the spacecraft’s data transmitter to beam radio waves into the dark regions near the poles and listen to their reflected echoes on the enormous (70 m) dish antenna of NASA’s Deep Space Network.  Interestingly, the reflections indicated an enhancement of “same sense” polarization within the (very large) resolution cell that contained Shackleton crater.  A collect of data from a nearby sunlit area (taken as an experimental control) did not show this peak.  The Clementine team interpreted the RF peak as evidence for the presence of a few percent water ice within the dark, cold interior of Shackleton crater.  The media quickly spread the startling news about water on our “bone-dry” Moon.

Such a controversial conclusion did not go unchallenged.  Some in the radar community argued that abundant wavelength-sized rocks on the surface were the source of the enhanced same sense reflection.  Since the lunar surface is indeed rocky, this interpretation could not be ruled out.  Then a few years later, the Lunar Prospector (LP) mission found an enhancement of hydrogen concentration at both poles of the Moon; as hydrogen is a major constituent of water, the idea ice exists in the dark areas gained credence and has lead to a decade-long scientific search (using a variety of techniques) for lunar polar ice.  Though many areas near the poles were studied in detail, attention continued to be drawn back to Shackleton and the area near the south pole.

From studying Clementine images, we discovered that part of the rim crest of Shackleton is one of the most sunlit areas on the Moon.  Now we had a double-attraction: constant sunlight with water ice nearby.  At a press briefing in 1996, I called this area of water and sunlight “the most valuable piece of real estate in the Solar System.” Nothing found subsequently has changed my mind on that judgment.

So what have we learned about Shackleton lately?  Many different, new sensors have flown to the Moon in the last few years, including radar, ultraviolet (UV) imaging, laser reflections, and low-light level imaging.  And yet again, Shackleton crater continues to confound us with contradictory evidence, both for and against the presence of water ice in its interior.

In 2009, the question regarding the presence of water ice somewhere near the lunar south pole was answered when the LCROSS impactor threw up a cloud of water vapor and ice particles during its collision with the floor of the nearby crater Cabaeus.  Spectral mapping instruments on three different spacecraft (Chandrayaan-1, Cassini, and EPOXI) documented the presence of adsorbed water on the lunar surface, increasing in concentration with latitude toward both poles.  A small impact probe flown by India (MIP) passed through a water vapor zone in the exosphere just above the lunar south pole.  And radar images from Mini-RF, our radar imaging experiment on both Chandrayaan-1 and Lunar Reconnaissance Orbiter (LRO), found evidence of high same sense reflections (just as Clementine had suggested in 1994) within the interior of Shackleton crater.  These new lines of supporting evidence were countered by Japanese researchers, whose Kaguya spacecraft imaged the interior of the crater and found morphology similar to other lunar craters in the same size-class.  But no one had ever claimed that the interior of Shackleton was a skating rink of pure ice – the lunar polar ice is partly covered by waterless dust and mixed with an unknown amount of dry regolith.

Interpretation of the new data continues to vex us.  The LOLA (laser altimeter) team on LRO recently published a paper that documents the high reflectivity (at 1 micron wavelength) of the walls of Shackleton.  Although the team’s favored interpretation is that this is caused by a constant exposure of fresh material on a steep slope, they also note that it is consistent with the presence of water ice on the walls of the crater.  In addition, a team analyzing neutron spectrometer data from both LP and LRO found evidence in the fast neutron data (never before analyzed) that water in the interior of Shackleton is a possible explanation for its signal.  Detailed analysis of the Mini-RF data for Shackleton corrected for its steep wall slopes and found that the presence of 5-10 wt.% water there provides the best model fit to the observed data.  Newly obtained UV images from LRO show the existence of water frost in the interiors of the craters Haworth and Shackleton, and the neutron detector on LRO shows enhanced hydrogen within both Shoemaker and Shackleton craters.  The Japanese team from Kaguya continue to insist that the no-ice interpretation is the correct one.

So we are left with a mystery.  Some evidence is pro-ice and some is contra-ice.  I find it interesting that for most of the investigators, new data does not necessarily change any minds, but tends to be interpreted in a way most favorable to their previously published ideas.  This should not be terribly surprising; the people who have argued for some specific interpretation presumably did so for good reasons and desire hard and clear-cut evidence to the contrary before abandoning a previously held position, one no doubt reached after much thought and soul-searching.

The way to unravel the water-ice mystery is to go to the surface of the lunar south pole (or both poles) and measure the composition of the surfaces in question.  Getting a definitive answer about the nature of lunar water would be game changing.   Some say the bigger mystery is:  Why hasn’t the United States sent a rover to the south pole of the Moon to take a closer look?

Composite image of the north pole of Mercury. Red are the areas of permanent shadow; yellow delineates radar bright deposits mapped from Earth. Data are plotted on a photomosaic of MESSENGER images. NASA

Mercury – the planet, not the element – was in the news this past week.  For some time, we had suspected that the poles of Mercury might harbor deposits of water ice.  This – on a planet so close to the Sun that the surface temperature at the equator is hot enough to melt lead!

Yet like the Moon, Mercury’s spin axis is perpendicular to the plane in which it orbits the Sun.  This means that large craters near Mercury’s poles lie in permanent shadow (“shivering” around -170° C), unaffected by the Sun’s searing heat (equivalent to more than eleven times the solar flux we get on Earth).  As on the Moon, these permanently shadowed areas get heat from only two sources – the 3 K background heat of space, created during the Big Bang some 15 billion years ago, and whatever heat is being generated now from the deep interior (a quantity that geophysicists call the heat flow of a planet).

Large planets (like Earth) generate heat mostly from the decay of radioactive elements deep inside them.  This heat is lost largely through the phenomenon of volcanism, in which melted rock from the interior is erupted onto a planet’s surface as lava and ash.  Smaller planets and moons likewise experience this heating and volcanism, but because they are have lower overall contents of heat-producing elements, their volcanic episodes occurred in the distant past.  Much of the heat of these smaller planets has been largely dissipated.  Thus, on Mercury, we suspect that the overall heat flow is very low, resulting in extremely cold temperatures on the floors of its permanently shaded polar craters.

For many years, astronomers have studied Mercury with radio telescopes from Earth (using radar to make images of its surface).  Because the orbital inclination of Mercury is relatively high (about 7°), we can get a fairly good look into the interiors of the polar craters.  Interestingly, even though Mercury is much farther away than the Moon, we can see more of the mercurian polar areas because of this relatively high orbital inclination (the Moon’s orbital plane is inclined only 5°).  These radar pictures showed an amazing and unexpected feature – the dark areas are filled with material that is highly reflective at radio frequencies, properties similar to the surfaces of the icy moons of Jupiter (Europa, Ganymede and Callisto).

These results were so unexpected and startling that debate raged for many years whether these deposits really were what they appeared to be: water ice.  Facts are stubborn things and few materials have radio properties similar to ice.  Some suggested that sulfur might be an alternative explanation, but provided little evidence for such behavior.  Moreover, another moon of Jupiter, Io, which has a surface largely composed of sulfur, does not show the radar brightness or “glint” seen on the other, ice-rich Jovian moons.

The debate on the nature of the Mercury polar deposits has now been settled with the release of new data from the MESSENGER mission.  Launched on August 3, 2004, with insertion into obit around the planet on March 18, 2011, the spacecraft has been taking pictures and making measurements of Mercury for the last two years.  We have mapped the extent of darkness near the poles, measured the temperatures of the surface inside these regions, and detected the presence of significant amounts of hydrogen there.  All of these results are strongly supportive of the water ice interpretation.

The existence of ice near the poles of Mercury supports the case for water ice on our own Moon, although there are some significant differences between the two occurrences.  Like Mercury, the Moon’s spin axis is nearly perpendicular to the plane of its orbit around the Sun.  The similarity of the terrain of both bodies results in deep holes that hide large expanses of terrain from the glare and heat of the Sun.  Both objects have been volcanically active in the past, but not today, meaning that the average rates of heat flow on both are low.  These properties result in the creation of polar “cold traps” in which any entering volatile substance (such as water molecules) cannot escape.

The solid bodies of the inner Solar System are constantly hit by debris from comets and asteroids.  This material contains water, both in free form and bound within hydrous minerals.  On smaller objects (like the Moon and Mercury), most of this water is lost to space, but we suspected that some of it might be retained within these dark cold traps near the poles.  Now we know that such a process does occur.

Differences between the Moon and Mercury result in differing amounts and settings for their polar deposits.  Being much closer to the Sun, one might expect Mercury to contain less water ice, but a variety of evidence suggests that the opposite is the case.  The polar ice of Mercury appears to be greater in extent and thickness than comparable deposits on the Moon.  This probably results from two factors.  First, Mercury is a bigger object, with a surface gravity about twice that of the Moon.  Thus, it is more difficult for water to “escape” from Mercury.  Second, the closeness of Mercury to the Sun (the edge of biggest gravity well of the Solar System) results in a higher flux of cometary impacts there than experienced in the Earth-Moon system.  So more water is being added to Mercury, where it is more easily retained.

Nonetheless, both Moon and Mercury have similar polar environments and processes.  The long debate – a scientific controversy for over 50 years – about water at the poles of these objects has been resolved.  The next steps will be to characterize these deposits in situ using a soft lander and selected instruments to measure the amounts, states and distributions of water in the polar areas.  Because of the great difficulty in even getting into orbit around Mercury (let alone landing there), doing this first on the Moon will mostly likely happen first.  So, here again is another rationale for sending a robotic surveying lander and rover mission to the poles of the Moon – in addition to characterizing these areas for our future presence there, by inference, we will also learn about the polar processes on and environment of Mercury.

A planetary “two-fer.”  Let’s get on with it.