MOON

Introduction

The Moon, the natural satellite of Earth, has positively affected our development in many profound ways. Its orbital presence helps stabilize Earth’s axial precession and thus, prevents the alternating extremes of climate that some planets,(the time it takes to rotate once on its spin axis) is about 29 Earth days or 708 hours, and daylight hours on the Moon (sunrise to sunset) last almost 2 weeks. The Moon is famous for its low gravity, about one-sixth of Earth’s. Thus, an astronaut who weighs 200 pounds on Earth weighs only 34 pounds on the Moon.

The Moon moves in an elliptical path around Earth and completes its circuit once every 29 days. This time is equal to the amount of time it takes for the Moon to rotate once on its axis (the lunar day). In consequence, the Moon shows the same hemisphere (called the near side) to Earth at all times. Conversely, one hemisphere is forever turned away from us (the far side) (Fig. 1). Before the space age, the far side of the Moon was completely unknown territory, not revealed to human gaze until its face was first photographed by the Soviet spacecraft Luna 3 in 1959.
The elliptical orbit of the Moon results in a variable distance between Earth and Moon. At perigee (when the Moon is closest to Earth), the Moon is a mere 356,410 km away; at apogee (the farthest position), it is 406,697 km away. This is different enough so that the apparent size of the Moon in the sky varies; its average apparent size is a little smaller than that of a dime held at arm’s length. In works of art, a huge lunar disk looming above the horizon is often depicted, but such an appearance is an illusion. A Moon near the horizon can be compared in size with distant objects on the horizon, such as trees, making it seem large, and a Moon near zenith (overhead) cannot be compared easily with earthly objects, and hence, seems smaller.
The plane of the Moon’s orbit lies neither in the equatorial plane of Earth nor in ecliptic plane, in which nearly all the planets orbit the Sun (Fig. 2). This relation poses some constraints on models of lunar origin. The spin axis of the Moon is nearly perpendicular to the ecliptic plane; it has an inclination of about 1.5° from the vertical. This simple fact has some really significant consequences. Because its spin axis is vertical, the Moon experiences no ”seasons”, as does Earth, whose inclination is about 24°. So, as the Moon rotates on its axis, an observer at the pole would see the Sun hovering close to the horizon. A large peak near the pole might be in permanent sunlight, and a crater floor could exist in permanent shadow. In fact, we now know that such areas exist, particularly near the South Pole. The existence of such regions has important implications for a return to the Moon.

Figure 1. Index map of the Moon (Clementine albedo image), showing the location of some prominent lunar features. Apollo landing sites are shown by crosses.

Figure 2. Orbital planes and spin axes of Earth and Moon. Although Earth’s axis is tilted 23° from the ecliptic, the Moon’s is nearly perpendicular to it, resulting in grazing solar illumination near the poles.
As the Moon circles Earth, they occasionally block the Sun for each other, causing eclipses.A solar eclipse occurs when the Moon is between the Sun and Earth and can occur only at new Moon (the dayside of the Moon is facing the Sun). Because of the variable distance between Earth and the Moon, its inclined orbital plane, and the smaller size of the Moon, solar eclipses are quite rare (years may pass between total solar eclipses), so their occurrence is always subject to much hoopla. A lunar eclipse, in contrast, occurs when the Earth is between the Moon and the Sun. These events happen much more frequently, because Earth’s shadow has a much larger cross-sectional area than the Moon’s shadow. Lunar eclipses can occur only during a full Moon (or new Earth). As the shadow of Earth slowly covers the full Moon, it takes on a dull red glow, caused by the bending of some sunlight that illuminates the Moon through the thick atmosphere of Earth.
The Moon is gradually receding from Earth. Early in planetary history, Earth was spinning much faster, and the Moon orbited much closer than it does now. Over time, energy has been transferred from Earth to the Moon, causing the spin rate of Earth to decline and the Moon to speed up in its orbit, thus moving farther away (the current rate of recession is about 4 cm/year). Such recession will continue; some day, the Moon will be too far away to create a total solar eclipse! Fortunately for lovers of cosmic spectacles, this will not happen for at least another few million years.
As the Moon orbits Earth, we can peek around its edges because of a phenomenon known as libration. Libration in latitude is caused by the 7° inclination of the plane of the Moon’s orbit to Earth’s equator. This inclination allows us to ”look over the edge” of the Moon as it moves slightly above or slightly below the equatorial plane. Libration in longitude is caused by the Moon’s elliptical orbit, which permits Earth viewers to look around its leading or trailing edge. A small libration is also caused by parallax, which is the effect that allows you to see more by moving side to side, in this case by the diameter of Earth. All told, these libration effects permit us to see slightly more than a single hemisphere, and over time, we can see about 59% of the lunar surface.
The gravitational influence of Earth and Moon upon each other is considerable. Because of the gravitational tug of the Moon and Sun, the Earth experiences tides, which are bulges in the radius of Earth induced by gravitational attraction. Tides, it is often thought, are associated with the oceans, but solid Earth also undergoes an up and down motion caused by tides. Because Earth attracts the Moon just as much in reverse, the Moon also experiences a tidal bulge, one that mirrors the tidal effects on Earth. The raising and lowering of solid body tides on Earth and Moon causes friction inside the two planets, and this source of heat is called tidal dissipation. Such an energy source for planetary heat may have been very important early in the history of the solar system, when Moon and Earth were closer together, but it is currently only a minor source of heat.

Origin of the Moon and its Structure

In their surveys of the solar system, astronomers have discovered dozens of satellites around other planets. Yet, of the four inner planets, only Earth and Mars have moons (and Mars’ are probably captured asteroids). Ours is remarkably large as satellites go, particularly compared to the modest size of Earth itself. The creation of the Moon was thus an unusual event in terms of general planetary evolution, and our knowledge of the solar system—however detailed— would be profoundly incomplete without determining how our enigmatic satellite came to exist.
Traditionally, scientists have investigated three models of lunar origin. In the simplest hypothesis, termed co-accretion, Earth and Moon formed together from gas and dust in the primordial solar nebula and have existed as a pair from the outset. A second concept, called the capture scenario, envisions the Moon as a maverick world that strayed too near the Earth and became trapped in orbit— either intact or as ripped-apart fragments—due to our planet’s strong gravity. According to the third model, termed fission, the Earth initially had no satellite but somehow began to spin so fast that a large fraction of its mass tore away to create the Moon (1).
It was hoped that our astronauts would return with results that would allow us to choose decisively from among these three models. Study of the Apollo samples has provided some constraints on the true lunar origin, but none of these models has proven completely satisfactory. First, the Moon’s bulk composition appears to be similar, but not identical, to the composition of Earth’s upper mantle. Both are dominated by the iron- and magnesium-rich silicates pyroxene and olivine. But one important distinction is that, unlike Earth, the Moon generally lacks volatile elements. Another involves the relative dearth in lunar material of what are termed siderophile (”metal-loving”) elements such as cobalt and nickel, which tend to occur in mineral assemblages that contain metallic iron (Fig. 3).
A second key constraint comes from oxygen’s three natural isotopes: 16O, 17O, and 18O. Ratios of these isotopes are identical in lunar and terrestrial materials, which suggests strongly that the Moon and Earth originated in the same part of the solar system. These same ratios are different in meteorites such as eucrites (asteroidal basalts), the so-called SNC group (possible Martian igneous rocks), and various subgroups of ordinary chondrites.

Figure 3. Maps of the epithermal neutron flux near the South Pole of the Moon from the Lunar Prospector. Low epithermal flux indicates the presence of hydrogen. This map shows that the highest hydrogen concentrations are associated with areas of permanent darkness near the South Pole. In conjunction with positive radar evidence from Clementine, this indicates that water ice exists near the South Pole of the Moon.
Beyond geochemical evidence, several physical properties of the Earth-Moon system provide important clues in determining lunar origin. For example, the pair possesses a great deal of angular momentum. Also, the Moon’s orbit does not lie within the plane of Earth’s equator or of its orbit (the ecliptic plane). Finally, the Moon is gradually receding from Earth at roughly 3 cm per year—a curious effect caused by the gravitational coupling of the Moon and our oceans. Tidal bulges raised in seawater do not lie directly along the Earth-Moon line but actually precede it, because Earth’s rotation drags them along for some distance before they can adjust to the Moon’s changing location in the sky. This misalignment causes Earth’s rotation to decelerate slightly; the Moon in turn is pulled forward in its orbit, speeds up, and inches farther away. Unfortunately, we cannot determine the original Earth-Moon distance because the orbital recession going on now cannot be extrapolated back to the lunar origin.
A new idea for the birth of the Moon has gained popularity recently and even something of a consensus, although all the attendant problems that it poses have yet to be resolved. This idea is that a giant object, possibly a planet-sized body as big as Mars, hit Earth around 4.6 billion years ago (Fig. 4). It may have struck off-center, thereby increasing Earth’s rotational rate. A mixture of terrestrial and impactor material would have been thrown into Earth orbit and later coalesced to form the Moon.
Because this material would have jetted into space in a predominantly vaporized state, the giant-impact hypothesis could explain both the Moon’s dearth of volatile elements and its possible slight enrichment in refractory elements (those that remain solid at high temperature). To account as well for the Moon’s depletion in metallic iron and siderophile elements relative to Earth, theorists must assume that the incoming object had already differentiated into a core and mantle. Their calculations show that at least half to nearly all of the lunar mass was derived from the outer layers of the colliding body. So to create a proper Moon depleted in iron and siderophiles, these elements would have to be concentrated in the impactor’s core, which became incorporated into the Earth shortly after the initial collision. [See article Appollo 17 and the Moon by Harrison Schmitt elsewhere in this topic.]
The giant-impact hypothesis appears to explain, or allow for, several fundamental relations—not just bulk composition, but also the orientation and evolution of the lunar orbit. It also makes the uniqueness of the Earth-Moon system seem more plausible, that is, impacts of this magnitude might have occurred only rarely, rather than as a requirement for planetary formation. Part of the reason for this model’s current popularity is doubtless because we know too little to rule it out: key factors such as the impactor’s composition, the collision geometry, and the Moon’s initial orbit are all underdetermined.
Scientists realize that the advent of the giant-impact hypothesis has not ”solved” the problem of lunar origin. For example, the close genetic relation of Earth and Moon (inferred from the oxygen-isotope ratios) is not an obvious consequence of a giant impact, especially if most of the lunar mass derived from the projectile. Consequently, research into the effects of such cataclysmic impacts in early planetary history continues at a brisk pace. But this model of lunar origin appears to explain the most salient features of the Moon and has a minimum amount of special pleading.

Figure 4. Giant impact model of lunar origin. In this hypothesis, a Mars-sized planet hit the Earth at grazing incidence, throwing vaporized mantle into Earth orbit. This material later coalesced to form the Moon.

Scientific Results from Apollo

From 1969 to 1972, six Apollo expeditions set down on the Moon, allowing a dozen American astronauts to explore the lunar landscape and return with pieces of its surface (Fig. 1). The initial landing sites were chosen primarily on the basis of safety. Apollo 11 landed on the smooth plains ofMare Tranquillitatis, Apollo 12 on a mare site near the east edge of the vast Oceanus Procellarum. These first missions confirmed the volcanic nature of the maria and established their antiquity (older than 3 billion years). Later missions visited sites of increasing geologic complexity. Apollo 14 landed in highland terrain near the crater Fra Mauro, an area, it was thought, is covered by debris thrown out by the impact that formed the Imbrium basin. Apollo 15 was the first mission to employ a roving vehicle and the first sent to a site that contains both mare and highland units (the Hadley-Apennine region). Apollo 16 landed on a highland site near the rim of the Nectaris basin. The final lunar mission in the series, Apollo 17, was sent to a combination mare-highland site on the east edge of the Serenitatis basin (2).
The Soviet Union has acquired a small but important set of lunar samples of its own, thanks to three automated spacecraft that landed near the eastern limb of the Moon’s near side. Luna 16 visited Mare Fecunditatis in 1970, and Luna 24 went to Mare Crisium in 1976. A third site, in the highlands surrounding the Crisium basin, was visited by Luna 20 in 1972.
Altogether, these nine missions returned 382 kg of rocks and soil, the ”ground truth” that provides most of our detailed knowledge of the Moon. Though the most exhilarating discoveries came from studies completed years ago, today scientists around the world continue to examine these samples, establish their geologic contexts, and make inferences about the regional events that shaped their histories. What we have learned about the Moon’s three major surface materials—maria, terrae, and the soil-like regolith that covers both—is summarized in the following paragraphs.
Regolith. During the Moon’s history, micrometeorite bombardment thoroughly pulverized the surface rocks into a fine-grained, chaotic mass of material called the regolith (also informally called ”lunar soil,” though it contains no organic matter). The regolith consists of single mineral grains, rock fragments, and combinations of these that have been cemented by impact-generated glass. Because the Moon has no atmosphere, its soil is directly exposed to the high-speed solar wind, gases flowing out from the Sun that become implanted directly onto small surface grains. The regolith’s thickness depends on the age of the bedrock that underlies it and thus how long the surface has been exposed to meteoritic bombardment; the regolith in the maria is 2-8 meters thick, whereas in highland regions its thickness may exceed 15 meters.
The composition of the regolith closely resembles that of the local underlying bedrock. Some exotic components are always present, perhaps having arrived as debris flung from a large distant impact. But this is the exception rather than the rule. The contacts between mare and highland units appear sharp from lunar orbit, which suggests that relatively little material has been transported laterally. Thus, although mare regoliths may contain numerous ter-rae fragments, in general these derive not from far-away highland plateaus but are instead crustal material excavated locally from beneath the mare deposits.
Impacts energetic enough to form meter-size craters in the lunar regolith sometimes compact and weld the loose soil into a type of rock called regolith breccia. Once fused into a coherent mass, a regolith breccia no longer undergoes the fine-scale mixing and “gardening” taking place in the unconsolidated soil around it. Thus, regolith breccias are ”fossilized soils” that retain their ancient composition and also the chemical and isotopic properties of the solar wind from the era in which they formed.
Maria. Thanks to our lunar samples, there is no longer any doubt that the maria are volcanic. Mare rocks are basalts that have a fine-grained or even glassy crystalline structure (indicating that they cooled rapidly) and are rich in iron and magnesium. Basalts are a widespread volcanic rock on Earth, consisting mostly of the common silicates pyroxene and plagioclase, numerous accessory minerals, and sometimes olivine (an iron-magnesium silicate). But lunar basalts display some interesting departures from this basic formulation. For example, they are completely devoid of water—or any form of hydrated mineral—and contain few volatile elements in general. Basalts from Mare Tranquillitatis and Mare Serenitatis are remarkably abundant in titanium; sometimes they contain roughly 10 times more than is typically found in their terrestrial counterparts.
Mare basalts originated hundreds of kilometers deep within the Moon in the total absence of water and the near absence of free oxygen. There, the heat from decaying radioactive isotopes created zones of partially molten rock that ultimately forced its way to the surface. The occurrence of mare outpourings within impact basins is no chance coincidence, for the crust beneath these basins must have been fractured to great depth by the cataclysmic impacts that formed them. Much later, molten magmas rose to the surface through these fractures and erupted onto the basin floors.
Although they may appear otherwise, the maria average only a few hundred meters in thickness. These volcanic veneers tend to be thinner near the rims that confine them and thicker over the basins’ centers (as much as 2-4 km in some places). What the maria may lack in thickness they make up for in sheer mass, which frequently is great enough to deform the crust underneath them. This has stretched the outer edges of the maria (creating fault-like depressions called grabens) and compressed their interiors (creating raised ”wrinkle” ridges) (Fig. 5).
Basalts returned from the mare plains range in age from 3.8-3.1 billion years, a substantial interval of time. But small fragments of mare basalt found in highland breccias solidified even earlier—as long ago as 4.3 billion years. We do not have samples of the youngest mare basalts on the Moon, but stratigraphic evidence from high-resolution photographs suggests that some mare flows actually embay (and therefore postdate) young, rayed craters and thus may be no older than 1 billion years.
A variety of volcanic glasses—distinct from the ubiquitous, impact-generated glass beads in the regolith—were found in the soils at virtually all of the Apollo landing sites. They even were scattered about the terrae sites, far from the nearest mare. Some of these volcanic materials are similar in chemical composition, but not identical, to mare basalts and were apparently formed at roughly the same time.

Figure 5. Area in eastern Mare Imbrium, showing wrinkle ridges (left), representing compressional tectonics, and grabens (right), representing extensional tectonics. These two landforms are the typical tectonic features of the Moon.
One such sample, tiny beads of orange glass, came from the Apollo 17 site. They are akin to the small airborne droplets that accompany volcanic ”fire fountains” on Earth, like those in Hawaii. The force of the eruption throws bits of lava high into the air, and they solidify into tiny spherules before hitting the ground. The Moon’s volcanic glass beads have had a similar origin. The orange ones from the Apollo 17 site get their color from high titanium content, more than 9%, and some of them are coated with amorphous mounds of volatile elements like zinc, lead, sulfur, and chlorine (see Appollo 17 and the Moon in this volume). Terrae. One could easily imagine that the lunar highlands contain outcrops of the original lunar crust—much as we find in Earth’s continents. But what really awaited the astronauts was a landscape so totally pulverized that no traces of the original outer crust survived intact. Instead, most of the stones collected from the terrae were breccias, usually containing fragments from a wide variety of rock types that have been fused together by impact processes. Most of these consist of still older breccia fragments that attest to a long and protracted bombardment history.
The highland samples also include several fine-grained crystalline rocks that have a wide range of compositions. They are not breccias, but they were created during an impact. In these cases the shock and pressure were so overwhelming that the ”target” melted completely and created in effect entirely new rocks from whatever ended up in the molten mass. The impactors become part of this mixture, and these impact-melt rocks contain distinct elemental signatures of meteoritic material.
Based on the samples in hand, virtually all of the highlands’ breccias and impact melts formed between about 4.0 and 3.8 billion years ago (Fig. 6). The relative brevity of this interval surprised researchers—why were all the highland rocks so similar in age? Perhaps the rate of meteoritic bombardment on the Moon increased dramatically during that time. Alternatively, the narrow age range may merely the conclusion of an intense and continuous bombardment that began 4.6 billion years ago, the estimated time of lunar origin. To resolve the enigma, we must return to the Moon and sample its surface at carefully selected geologic sites.
A substantial number of small, whitish rock fragments found in the mare soils returned by Apollo 11 and 12 astronauts had compositions totally unlike that of basalts and virtually unmatched on Earth. They consisted almost entirely of plagioclase feldspar, a silicate rich in calcium and aluminum but depleted in heavier metals such as iron. A few prescient researchers postulated that these rocks came from the lunar highlands. The last four Apollo missions, sent to highland landing sites, confirmed that plagioclase feldspar dominates the lunar crust. The resulting implication was broad and profound: at some point in the distant past much of the Moon’s exterior—and perhaps its entire globe—had been molten.
The detailed nature of this waterless ”magma ocean” is only dimly perceived at present; for example, the lunar surface may not have been completely molten everywhere. But the consequences seem clear. In a deep, slowly cooling layer of lunar magma, crystals of low-density plagioclase feldspar would have risen upward after forming, and higher density minerals would have accumulated at lower levels. This segregation process, termed differentiation, left the young Moon with a crust that was, in effect, a low-density rock “froth” tens of kilometers thick that consists mostly of plagioclase feldspar. At the same time, denser minerals (particularly olivine and pyroxene) became concentrated in the mantle below—the future source region of mare basalts.

Figure 6. Histogram of lunar rock ages. Note that most lunar rocks are extremely ancient and date from 3-4.5 billion years (the solar system is 4.6 billion years). Virtually all of the impact melts from the lunar highlands date from the narrow time interval between 3.8 and 4.0 billion years ago, a time known as the lunar ”cataclysm.”
It is unclear to what depth the magma ocean extended, but the coatings of volatile elements discovered on some mare glasses provide an important clue. If the Moon’s exterior really was once molten, the most volatile components in the melt would have vaporized and escaped into space. But the volatile-coated glasses sprayed onto the lunar surface long after the magma ocean solidified. If the glasses’ compositions did not change in their upward migration from the lunar interior, they imply that volatile-rich pockets remained (and perhaps still exist) in the upper mantle. The implication, therefore, is that the magma ocean was at most only a few hundred kilometers deep (3).
The highland samples returned by the last four Apollo crews provided other surprises. Unlike glasses and basalts, which quench quickly after erupting onto the surface, some of the clasts in the highland breccias contained large, well-formed crystals, indicating that they had cooled and solidified slowly, deep inside the Moon. These igneous rocks sometimes occur as discrete specimens. At least two distinct magmas were involved in their formation. Rocks composed almost completely of plagioclase feldspar, that have just a hint of iron-rich silicates are called ferroan anorthosites. They are widespread in the highlands. Absolute dating of the anorthosites has proved difficult, but it appears that they are extremely ancient, having crystallized very soon after the Moon formed (4.6 to 4.5 billion years ago).
The highlands’ other dominant rock type is also abundant in plagioclase feldspar, but it contains substantial amounts of olivine and a variety of pyroxene low in calcium. This second class of rocks is collectively termed the Mg-suite,so called because they contain considerable magnesium (Mg). These rocks appear to have undergone the same intense impact processing as the anorthosites, and their crystallization ages vary widely—from about 4.3 billion years to almost the age of the Moon.
The anorthosite and Mg-suite rocks could not have crystallized from the same “parent” magma, so at least two (and probably more) deep-seated sources contributed to the formation of the early lunar crust. Conceivably, both magmas might have existed simultaneously during the first 300 million years of lunar history. This would contradict our notion of the Moon as a geologically simple world and greatly complicate our picture of the formation and early evolution of its crust.
During early study of the Apollo samples, an unusual chemical component was identified that is enriched in incompatible trace elements—those that do not fit well into the atomic structures of the common lunar minerals plagioclase, pyroxene, and olivine as molten rock cools and crystallizes. This element group includes potassium (K), rare-earth elements (REE) such as samarium, and phosphorus (P); geochemists refer to this element combination as KREEP. It is a component of many highland soils, breccias, and impact melts, yet the trace-element abundances remain remarkably constant wherever it is found. Moreover, its estimated age is consistently 4.35 billion years. These characteristics have led to the consensus that KREEP represents the final product of the crystallization of a global magma system that solidified aeons ago.
But the evidence of chemically distinct, widespread volcanic rocks in the highlands—KREEP-rich or otherwise—remains tenuous. Some highland rocks are compositionally similar to mare basalts yet exhibit KREEP’s trace-element concentrations. For example, the Apollo 15 astronauts returned with true basalts that probably derive from the nearby Apennine Bench Formation (Fig. 5), a large volcanic outflow situated along the Imbrium basin’s rim. These ”KREEP basalts” have a well-determined age of 3.85 billion years, so the Imbrium impact must have occurred before this date and probably just before the Apennine Bench Formation extruded onto the surface. Thus, although the extent and importance of highland volcanism remains unknown, it apparently took place early in lunar history and contributed at least some of the KREEP component observed in highland breccias and impact melts (4).

Water on the Moon

An abundant supply of water on the Moon would make establishing a self-sustaining lunar colony much more feasible and less expensive than presently thought. Study of lunar samples revealed that the interior of the Moon is essentially devoid of water, so no underground supplies could be used by lunar inhabitants. However, the lunar surface is bombarded with water-rich objects such as comets, and scientists have suspected that some of the water in these objects could migrate to permanently dark areas at the lunar poles and perhaps accumulate to usable quantities.
Water is constantly being added to the Moon from impacting comets and water-rich asteroids. Where would this water end up? Most of it would be split by sunlight into its constituent atoms of hydrogen and oxygen and lost into space, but some would migrate by literally hopping along to places where it is very cold. It was postulated that the polar regions might have areas that are permanently shadowed, hence permanently cold. The water might accumulate there.
The Moon’s axis of rotation is nearly perpendicular to the plane of its orbit around the Sun (Fig. 2). Although the plane of the Moon’s orbit about Earth is inclined about 5°, its equator is inclined about 6.5°, resulting in a 1.5° inclination of the Moon’s spin axis to its orbital plane around the Sun. This means that the Sun always appears close to the horizon at the poles of the Moon.
It has been calculated that temperatures in these permanently dark areas may be a scold as40to50 K (— 230° to — 220°C), only a few tens of degrees above absolute zero. Moreover, these ”cold traps” have existed on the Moon for at least the last 3-4 billion years—plenty of time to accumulate water from impacting comets.
To determine whether there is water on the Moon, we had to await the results of polar-orbiting, global mappers. Two missions, Clementine and Lunar Prospector, sent to the Moon in the 1990s, looked for evidence of water at the poles.

Results from Post-Apollo Robotic Missions

Clementine. Clementine was a mission designed to test the space worthiness of a variety of advanced sensors for use on military surveillance satellites and, at the same time, to gather useful scientific information on the composition and structure of the Moon and a near-Earth asteroid. Conducted jointly by the Ballistic Missile Defense Organization (BMDO, formerly the Strategic Defense Initiative Organization) of the U.S. Department of Defense and NASA, Clementine was sent for an extended stay in the vicinity of Earth’s Moon on 25 January 1994 and arrived at the Moon on 20 February 1994. The spacecraft started systematic mapping on 26 February 1994, completed mapping on 22 April 1994, and left lunar orbit on 3 May 1994 (5).
During 71 days in lunar orbit, Clementine systematically mapped the 38 million square kilometers of the Moon in 11 colors in the visible and near-infrared parts of the spectrum. In addition, the spacecraft took tens of thousands of high-resolution and mid-infrared thermal images, mapped the topography of the Moon with a laser ranging experiment, improved our knowledge of the surface gravity field of the Moon through radio tracking, and carried a charged particle telescope to characterize the solar and magnetospheric energetic particle environment. We have had our first view of the global color of the Moon, identifying major compositional provinces; studied several complex regions, mapping their geology and composition in detail; measured the topography of large, ancient impact features, including the largest (2500 km diameter), deepest (more than 10 km) impact basin known in the solar system; and deciphered the gravitational structure of a young basin on the limb of the Moon, finding that a huge plug of the lunar mantle is uplifted below its surface.
The color of the Moon in the visible to near-infrared part of the spectrum is sensitive to variations in both the composition of surface material and the amount of time that material has been exposed to space. The Clementine filters were selected to characterize the broad lunar continuum and to sample parts of the spectrum that, it is known, contain absorption bands diagnostic of iron-bearing minerals. By combining information obtained through several filters, multispectral image data are used to map the distribution of rock and soil types on the Moon.
Preliminary studies of areas of already known geologic complexity allow us to identify and map the diversity within and between geologic units, which have both impact and volcanic origins. The Aristarchus Plateau is a rectangular, elevated crustal block about 200 km across, surrounded by the vast mare lava plains of Oceanus Procellarum. Clementine altimetry shows that the plateau is a tilted slab that slopes down to the northwest and rises more than 2 km above Oceanus Procellarum on its southeastern margin. The plateau was probably uplifted, tilted, and fractured by the Imbrium basin impact, which also deposited hummocky ejecta on the plateau surface.
The plateau has experienced intense volcanic activity, both effusive and explosive. It includes the densest concentration of lunar sinuous rilles, including the largest known, Vallis Schroteri, which is about 160 km long, up to 11km wide, and 1 km deep. The rilles in this area begin at ”cobra-head” craters, which are the apparent vents for low-viscosity lavas that formed the rilles. These and other volcanic craters may have been the vents for a ”dark mantling” deposit that covers the plateau and nearby areas to the north and east. This dark mantling deposit probably consists primarily of iron-rich glass spheres (pyroclastics or cinders) and has a deep red color. Rather than forming cinder cones as on Earth, the lower gravity and vacuum of the Moon allows the pyroclastics to travel much greater heights and distances and thus deposit an extensive regional blanket.
The Aristarchus impact occurred relatively recently in geologic time, after the Copernicus impact but before the Tycho impact. The 42-km diameter crater and its ejecta are especially interesting because of their location on the uplifted southeastern corner of the Aristarchus plateau. As a result, the crater ejecta reveal two different stratigraphic sequences: that of the plateau to the northwest and that of a portion of Oceanus Procellarum to the southeast. This asymmetry is apparent in the colors of the ejecta, which are reddish to the southeast, dominated by excavated mare lava, and bluish to the northwest, caused by the excavation of highlands materials in the plateau. The extent of the continuous ejecta blanket also appears asymmetrical: it extends about twice as far to the north and east than in other directions, approximately following the plateau margins. These ejecta lobes could be caused by an oblique impact from the southeast, or it may reflect the presence of the plateau during ejecta emplacement.
The Clementine multispectral data will enable us to reconstruct the three-dimensional composition and geologic history of this region. In this color-ratio composite, fresh highlands materials are blue, fresh mare materials are yellowish, and mature mare soils are purplish or reddish. The subsurface compositions, buried beneath a few meters or tens of meters of pyroclastics or Aristarchus ejecta, are revealed by craters that penetrated the surface layers and by steep slopes such as those along the walls of the rilles. From this mosaic, we have seen that the plateau is composed ofa complex mixture of materials, but that the rilles formed primarily in lavas, except for the cobra-head crater of Vallis Schroteri that formed in highland materials.
The laser ranging data from Clementine has allowed us to see the large-scale topography (or relief) of the lunar surface on a nearly global basis (Fig. 7). A striking result from this experiment is confirmation of the existence of a population of very ancient, nearly obliterated impact basins, randomly distributed across the Moon. These basins had been postulated on the basis of obscure circular patterns on poor quality photographs; Clementine laser ranging has provided dramatic confirmation of their existence, including their surprising depth, ranging from 5-7 kilometers, even for the most degraded features. Gravity data obtained from radio tracking of Clementine indicates that these great holes in the Moon’s crust are compensated for by plugs of dense rocks far below the surface; such dense rocks are probably caused by structural uplift of the mantle (the iron- and magnesium-rich layer below the low-density, aluminum-rich crust) beneath these impact basins. Finally, Clementine laser ranging data have shown us the dimensions of the largest confirmed basin on the Moon, the 2500-km diameter South Pole-Aitken basin (Fig. 7): this feature averages more than 12 kilometers deep that makes it the largest, deepest impact crater known in the solar system (6).
Although the Clementine spacecraft did not carry instruments designed to look for lunar ice, during the mission, we improvised an experiment that allowed us to address this question. Radio waves are reflected from planetary surfaces differently, depending on the compositional makeup of those surfaces. Specifically, radio waves are scattered in all directions by reflection from surfaces made up of ground-up rock (as are the terrestrial planets, which include most of the Moon, Mercury, Venus, Mars, and the asteroids), whereas radio waves are reflected more coherently from ice surfaces (the polar caps of Mercury and Mars and the icy surfaces of Jupiter’s satellites Europa, Ganymede, and Callisto). When radio waves encounter ice, it is partly absorbed and reflected multiple times by internal flaws in the ice, then reflected back out into space. A consequence of multiple reflections is that some of the radio reflections come back in the same sense as they were transmitted (think of the reflection of light from two mirrors—reflection from a single mirror makes text unreadable, but double reflection makes the text ”normal” again.) Thus, ice reflects back partly in the same sense as incident waves (5).

Figure 7. Global map of the topography of the Moon derived from Clementine laser altimetry and stereo photography (for the poles). The Moon shows a huge dynamic range of elevations, mostly caused by the presence of large, impact craters, called basins. Note the huge depression on the southern far side—this is the South Pole-Aitken basin, more than 2500 km in diameter and more than 12 km deep, the largest, deepest impact crater known in the solar system.
Analysis of the data returned from this radio-wave experiment performed in 1994 while the Clementine spacecraft was orbiting the Moon reveals that deposits of ice exist in permanently dark regions near the South Pole of the Moon. Initial estimates suggest that ice exists at both poles and has the volume of a small lake more than 1 billion cubic meters (~ 1km3). This amount of water would be equivalent to the fuel (hydrogen and oxygen) used for more than 100,000 launches of the Space Shuttle.
Lunar Prospector. The Lunar Prospector (LP), the first of NASA’s new, cheaper, “Discovery”-class missions, was launched to the Moon on 6 January 1998. Prospector orbited the Moon in a 100-km altitude, polar orbit for more than 18 months. It carried a variety of instruments that, in many ways, complemented the instruments of the earlier Clementine mission. In part, Lunar Prospector’s stated objectives were to map the resources of the Moon, including assessments of polar volatiles and basic elemental composition. In addition, it would map the gravity and magnetic fields of the Moon during its 1-year nominal mission.
The spacecraft was a spin-stabilized microsat, about 100 kg in dry mass. It carried a gamma-ray spectrometer, designed to measure lunar surface chemical composition, a neutron spectrometer (to measure hydrogen in the regolith and to search for ice in the polar regions), a magnetometer, and radio tracking for gravity field measurements. Because all instruments were non pointing and had 4p fields of view, the resolution of compositional maps is fixed by the orbital altitude. For the gamma-ray and neutron maps, nominal surface resolution is about 100 km. Because the orbital altitude was lowered during the extended mission, some higher resolution data (about 30 km) are also available (Fig. 3).
Of particular interest in the LP data is the distribution of the element thorium (Th) on the Moon. This element tracks the distribution of KREEP (mentioned before). Most of the Th in the upper crust of the Moon is highly concentrated in a large, regional oval centered on Oceanus Procellarum; smaller concentrations are observed in the floor of SPA basin. The Procellarum Th oval is unexplained. It may represent an original heterogeneity in the crust of the Moon (inherited from global differentiation) or, it could be the result of material excavated and thrown across the surface by the impact that created the Imbrium basin. The (lower in magnitude) Th anomaly in the SPA basin floor suggests that the enrichment in Th here may be the result of the exposure of lower crustal material.
Lunar Prospector’s neutron spectrometer detected high concentrations of hydrogen at both poles (Fig. 3). In the form of water ice, the latest results from LP show an amount of hydrogen equivalent to about 10 km3 of ice; the South Pole has slightly more than the North Pole (contradicting early analyses that suggested more at the North Pole). Moreover, the low-altitude (high-resolution) neutron data show that these high concentrations of hydrogen are correlated with the large areas of darkness in the Clementine mosaics (Fig. 3). This result almost certainly means that large quantities of water ice exist in these dark areas, confirming the earlier result of the Clementine bistatic radar experiment.
Measurements of the magnetic field of the Moon from LP has confirmed that the Moon possesses a small, metallic core, about 400 km in diameter, or, about 2% of the mass of the Moon. This core is mostly iron but may contain significant amounts of iron sulfide (FeS). Numerous, intense zones of magnetism are associated with bright swirl material on the Moon. LP found that some of these magnetic anomalies are intense enough actually to deflect the solar wind from the surface. If such fields are geologically old, there should be enhanced solar wind gas implantation along the margins of these anomalies and shielded areas beneath the magnetic ”bubbles.”
At the end of its mapping mission, the LP spacecraft was deliberately crashed into the Moon, near the South Pole, in hopes that a cloud of water vapor might be released, which could then be seen by telescopes on Earth. This experiment was conducted on 31 July 1999; no vapor cloud was detected. This negative result does not mean there is no water ice on the Moon; it only means that we did not detect it in this experiment.

Future Human Activities on the Moon

The discovery of ice on the Moon has enormous implications for a permanent human return to the Moon. Water ice is made up of hydrogen and oxygen, two elements vital to human life and space operations. Lunar ice could be mined and disassociated into hydrogen and oxygen by electric power provided by solar panels deployed in nearby illuminated areas or by a nuclear generator. This hydrogen and oxygen is a prime rocket fuel, giving us the ability to refuel rockets at a lunar ”filling station” and making transport to and from the Moon more economical by at least a factor of 10. Additionally, the water from lunar polar ice and oxygen generated from the ice could support a permanent facility or outpost on the Moon. The discovery of this material, rare on the Moon but so vital to human life and operations in space, will make our expansion into the solar system easier and reaffirms the immense value of our own Moon as the stepping-stone into the Universe.
The Moon as a Planetary Touchstone. Beyond obtaining new samples, emplacing a global network of geophysical stations would help us learn more about the Moon’s mantle and core structure, variations in its crustal thickness, and the enigmatic lunar paleomagnetism. It would also lead to a more accurate determination of the enrichment of the Moon in refractory elements by measuring lunar heat flow (because two of these elements, uranium and thorium, are radioactive and thus important sources of heat).
Eventually, humans will probably go to the Moon to live, and establishing a permanent presence there opens up scientific vistas that are difficult to foresee clearly. Each Apollo mission provided some geologic surprise within its sample collection. So there is little doubt that both the variety of rock types and geologic processes that have operated on the Moon exceed by far those that we have currently deciphered. From a permanent lunar base, we could begin a detailed exploration of our complex and fascinating satellite that could last for centuries — and uncover its secrets and also the early history of our home planet as well.
As fascinating as it is, our views on the evolution and history of the Moon have more relevance than just to lunar study. During the last 20 years, we made our first exploration of the planets. We surveyed and photographed all of the terrestrial planets, Mercury, Venus, and Mars, and conducted our initial reconnaissance of the rocky and icy satellites of the giant outer planets. We landed robot spacecraft on Mars and analyzed its surface materials. All of the planetary bodies studied to date show, to different degrees, the same kinds of surface and geologic processes first recognized and described on the Moon. Much of our understanding of planetary processes and history comes by comparing surface features and environments among the planets. In any such comparison, reference is inevitably made to knowledge we obtained from lunar exploration.
One of the most startling results from Apollo was the concept of the magma ocean. The Moon is a relatively small object, transitional in size between planets and asteroids. In general, the amount of heat that a planetary object contains is related to its size; larger planets contain more heat-producing elements. If a body as small as the Moon could undergo global melting, it is a near certainty that the other terrestrial planets also melted. The idea that early Earth underwent global melting has been bandied about for many years; the evidence of the magma ocean made such speculation respectable. Now, we think that early planetary melting may be a widespread phenomenon and could be responsible for creating all of the original crusts of the planets.
Knowing that global melting occurred is one thing, understanding how it operated in detail is another task altogether. The Moon is a natural laboratory to study this process. One of the most fundamental discoveries of Clementine is that the aluminum-rich, anorthosite crust is indeed global and provides strong support for the magma ocean. Our next task is to understand the complex processes at work in such an ocean. Did a “chilled” crust form and if so, are any pieces of it left? Such material would allow us to determine directly the bulk composition of the Moon, a parameter that is estimated indirectly (and very imprecisely) now. Are there any highly magnesian rocks in the highlands that are related to the magma ocean, not to the younger magnesium-rich suite of rocks? We have searched the Apollo collections for such rocks but have found none. They may exist at unvisited sites on the Moon.
Another process common to all of the planets is volcanism. The Moon is the premier locality to study planetary volcanism. The flood lavas of the maria span more than a billion years of planetary history and probably come from many different depths within the Moon. Thus, the lava flows are actually probes of the interior of the Moon, both laterally (across its face) and vertically (throughout the depths of the mantle). The inventory and study of the mare basalts will allow us to categorize both of these dimensions and the important additional dimension of time. By sampling, chemically analyzing, and dating many different samples of lava that cover the globe, we can piece together the changing conditions of the deep mantle across long periods.
The styles of eruption responsible for the maria appear to be typical of those on other planets. Flood volcanism—the very high rates of effusion responsible for the mare lavas—is seen on every terrestrial planet and appears to be especially widespread on Mars and Venus. What is largely unknown is the size and shape of the vents through which these lavas were extruded. On the Moon, eruptive vents might be exposed in several locations, including within the walls of grabens and irregular source craters. Detailed exploration and study of such features would help us understand a style of volcanism ubiquitous on the planets. Small, domelike volcanoes, such as found in the Marius Hills, can be explored and examined to understand the styles and rates of lava eruption in creating these features. Small shield volcanoes are common on the surfaces of Mars and, particularly, on Venus. Relatively exotic processes, such as the erosion of terrain by flowing lava, has been proposed for the sinuous rilles of the Moon. Study of large rilles could help us decide whether this concept is correct.
From studying the Moon, we know that impact is one of the most fundamental of all geologic processes. Based on its population of craters of all sizes,where better to study and understand this important shaper of surfaces than on the Moon? Our ignorance is particularly vast for craters at the larger end of the size spectrum. Craters such as Copernicus (100 km in diameter) offer a window into the upper crust through the study of their ejecta, and a view of the middle level of the crust through their central peaks (which have uplifted rocks from 15-20 km deep). The ubiquity of craters that have such central peaks allows us to reconstruct the nature of the crust in detail. Studying large craters will also clarify the nature of the process of impact. We suspect that large craters grow proportionally, that is, they excavate amounts of material to depths that can be predicted from studying smaller craters. But we are not certain that this is true. By studying craters on the Moon, we can determine whether this pattern of growth behaves as predicted.
The giant basins of the Moon pose many mysteries. Understanding such craters is important because we have found basins on the other planets, particularly on Mars and Mercury. Basins form in the earliest stages of planetary history. They excavate and redistribute the crust, serve as depressions where other geologic units may be deposited (such as stacks of thick lava), and may trigger the eruption of massive floods of lava. Yet for all of their importance, we still do not fully understand how far or how deep excavation extends, how the multiple, concentric rings are formed, how the ejected material behaves, and how far ejecta gets thrown as a function of mass. The Moon has preserved more than 40 of these important features for our study, all in various states of preservation and all dating from the very earliest phase of planetary history. Here we can study the process of large-body impact better than anywhere else in the solar system.
The temporal record of impacts in the Earth-Moon system can also be read on the lunar surface. On the basis of evidence of mass extinctions on Earth and from the ages of impact melts, cycles of bombardment and an early impact “cataclysm” have been proposed. Neither of these ideas have been proven, but both are potentially revolutionary and make us look at the history of the planets in a new way. The evidence to test these two ideas lies on the Moon. Episodic bombardment can be tested by sampling the melt sheets of many different craters and dating the samples. Episodes of intense cratering will be evident if groups of melt rocks have the same ages, spaced at constant intervals. On the other hand, a continuous distribution of ages would argue that such episodes of intense crate-ring do not occur. We cannot conduct this experiment on Earth or on any other planet—this highlights the uniqueness of the Moon for answering many questions in planetary science, questions that have application to a host of other scientific fields. The cataclysm is important because if the Moon underwent such an unusual bombardment history, Earth may also have experienced it, and applying the inferences made from lunar cratering to the other terrestrial planets would have to be reevaluated.
Studying the regolith will be one of the most important tasks during a lunar return. The regolith contains exotic samples flung from rock units hundreds of kilometers away. Using the regolith as a sampling tool, we can conduct a comprehensive inventory of the regional rock units by collecting many samples from the regolith at a single site. The regolith also contains a record of the output of the Sun during the last 3 billion years. To read this record, we must understand how the regolith grows and evolves. This knowledge will come only when we can study the regolith and its underlying bedrock in detail; to learn how its layers are formed; how the soil is exposed, buried, and re-exposed; and how volatile components might be mobilized and migrate through the soil. This knowledge is essential if we are to realize the goal of using the regolith as a recorder of the solar and galactic particles that have struck the Moon during its history. Astronomy from the Moon. One day, the Moon will become humanity’s premier astronomical observing facility. Consider its advantages. The Moon rotates very slowly (once every 29 days), so its ”nighttime” is 2 weeks long. Moreover, because the Moon has no atmosphere, we can observe stars constantly, even in the daytime! The lack of an atmosphere also means that telescopes on the Moon will not be plagued by the ”blurring” that a turbulent, thermally unstable air layer causes and that observations will not be degraded by ”light pollution,” the airglow that interferes with astronomy on Earth. The vacuum of the Moon also means that there are no ”absorptions” to prevent certain wavelengths of radiation from being observed, such as the infamous ”water absorption” in the atmosphere of Earth. The surface materials could be used as construction material for observatories.
Telescopes in Earth orbit or elsewhere in deep space also have many of these advantages. Why is the Moon better than these localities? The principal reason is that the Moon provides a quiet, stable platform. Seismic activity on the Moon is roughly one million times less than that of Earth. Because the Moon is a primitive, geologically dead world, it does not have the shifting, massive plates of our own dynamic Earth, along with its associated seismic trembling. Such stability of the surface would allow us to construct extremely sensitive instruments for observation, instruments that could not be constructed on Earth.
Likewise, a space-based telescope, such as the Hubble Space Telescope, must have its attitude carefully stabilized to achieve high-resolution observations. In addition, space telescopes have stringent pointing requirements and must not be pointed anywhere near the Sun. Both of these drawbacks mean that a telescope free-flying in space must carry attitude control fuel, precision gyroscopes, and equipment to protect the telescope’s optics from solar burn damage. On the Moon, the quiet, stable base of the surface would alleviate such problems, allowing sensitive instruments to be erected and operated easily (7).
One such instrument, an interferometer, consists of an array of smaller telescopes. The smallest object that a telescope can see clearly is directly related to the size of its aperture, or the diameter of its mirror or lens. A telescope that has a larger aperture can resolve smaller or more distant features than a telescope that has a smaller aperture. However, there is a practical limit to the size that we can make telescopes. After a certain size is attained, such instruments become unwieldy and unstable. Interferometry is a technique whereby a series of small telescopes are operated as a larger aperture instrument. Each element of the array images some distant object. The light waves from this image are ”added” in perfect phase and frequency to identical images obtained from other telescopes in the array, each separated by as much as several kilometers. The effect of this addition is to create an image of the same quality as if a telescope that had an aperture size equal to the separation distance had been used to image the star. This means that we can construct ”telescopes” whose effective aperture sizes are kilometers across!
Even a small interferometer on the Moon would exceed the resolving capabilities of the very best telescopes on Earth and could even surpass the capabilities of the Hubble Space Telescope. Using such an instrument, we could resolve the disks of distant stars and observe and catalog ”star spots,” which are clues to the internal workings of stars. We could see individual stars in distant galaxies and catalog the stellar makeup of a variety of galaxy types. Optical interferometers could look into a variety of nebulae and observe the details of new stars and stellar systems in the very act of formation. Such a window onto the Universe is very likely to revolutionize astronomy in the same way it was changed when Galileo turned his crude ”spyglass” toward the heavens in 1609.
The field of planetary astronomy would be completely changed by lunar observatories. The incredible resolving power of these instruments would allow us to examine deep sky systems, resolve the disks of extrasolar planets, and catalog the variety possible in other solar systems of our galaxy. All of our concepts of the way planets are created and evolve are derived from a single example, our own solar system. By observing the vast array of planetary systems circling nearby stars, we could see how they differ in such aspects as the number and spacing of planets, the ratio of giant gas planets to rocky ”terrestrial” objects, and the evolution of those individual planets. Spectroscopic observations of these planets would allow us to determine the composition of their atmospheres, if any, and the surface composition, if visible. The compositions of planetary atmospheres could indicate the presence of life on these bodies. Carbon dioxide mixed with free oxygen in a planetary atmosphere would be a telltale indication of plant life, which ”breathes” the former and manufactures the latter.
Astronomers look at the sky in many wavelengths other than the optical band. High-energy regions of the spectrum, such as X-ray and gamma-ray radiation, also contain important information about processes that occur in stars and galaxies. Supernovas (the sudden explosion of certain stars) produce copious amounts of high-energy radiation and energetic particles, such as cosmic rays. We think that certain particles derived from supernova explosions predate our solar system and that these stellar eruptions can induce planetary formation. We have already mentioned the use of the regolith as a recorder of energetic particle events. Using a high energy lunar observatory, we can watch the effusion of these particles and radiation as they happen (supernovas are common and typically are occurring in some part of the sky at any given time). We have only begun to observe such stellar explosions from space and no doubt have much to learn.
At the other end of the spectrum, the Moon is an ideal place for observation in the thermal infrared and radio bands. Observing the sky in the long wavelengths of the thermal infrared (10 microns and longer wavelengths) is difficult because such detectors measure heat and they must be cooled to very low temperatures for use. Usually, this is done at great cost and difficulty by using cryogenic gases, such as very cold liquid helium (— 300°C). Preserving such cold temperatures requires a lot of electrical power. The Moon is naturally cold. Surface temperatures during the lunar night may become as low as — 160°C. In the shadowed areas near the South Pole (Fig. 3), it may be as cold as — 230°C, only 40 K above absolute zero! These temperatures would permit passive cooling of infrared detectors, allowing telescopes to be operated without costly and difficult-to-use cryogenic, cooling gas. Observing the thermal infrared sky would tell us much about dust clouds and nebulae in which new stars and planets are being formed.
The sky at certain radio wavelengths is almost completely unknown. Earth has an ionosphere, a layer of electrically charged atoms that makes certain radio waves bounce off it, and we cannot see the radio sky at certain frequencies. Moreover, the electrical din of Earth caused by radio stations, microwave cookers, automotive ignition systems, and the thousands of other static generators of modern civilization cause radio astronomers great vexation in their attempts to map the sky. The far side of the Moon is the only known place in the Universe that is permanently shielded from the radio noise of Earth. Locating a radio telescope on the far side would permanently place 3600 kilometers of solid rock between the observatory and the radio din of Earth. We will see sky for the first time at some radio wavelengths. History has shown us that whenever we look at the Universe with a new tool or through a new window of frequencies, we learn new things and re-examine present knowledge with new and sometimes startlingly different appreciation.
We can use the distinctive lunar terrain to our advantage. The small, bowl-shaped craters of the Moon are natural features that could be turned into gigantic “dish” radio antennas by laying conductive material (e.g., chicken wire) on their floors and hanging a receiver over and above the center of the crater at the “focus” of the dish. This technique has already been done on Earth at the famous Arecibo Radio Observatory in Puerto Rico, using a natural depression in the limestone bedrock to create a giant “dish” antenna. Interferometers could also be built at radio wavelengths, creating radio telescopes that have huge apertures. The large, flat mare plains would make an ideal site to lay out arrays of smaller telescopes. Manufacture of antenna elements from local resources could make constructing of extremely large instruments feasible.
Using the Moon as an astronomical observatory has great advantages, and many astronomers have taken up the banner for a return to the Moon. In the minds of some, astronomy is the principal reason for a lunar return. However, an observatory on the Moon also has its problems. The ubiquitous and highly abrasive dust must be very carefully controlled. Movements of people and machines will have to be minimized around telescope facilities because the slightest stirring up of dust could coat delicate optical surfaces. We would have to shield energetic detectors carefully from solar flares (this could be done by using the local regolith material.) We must guard against radio contamination of the far side because extensive operations of a base could ruin certain radio astronomical observations. Our task before a lunar return is to understand the impact of each problem fully and devise methods of working around it.
Despite these problems, the Moon offers unique opportunities for astronomy. Each time we see the sky more clearly or more completely, we obtain new insights into the way the universe works. A lunar window on the Universe around us will give us a new appreciation and understanding of both the Universe and of our place in it.