Exploring Shorty crater during the Apollo 17 mission to the Moon. This was the only Apollo mission to include a geologist. NASA photo.
The geology of the Moon (sometimes called selenology, although the latter term can refer more generally to "lunar science") is quite different from that of the Earth. The Moon lacks a significant atmosphere, which eliminates erosion due to weather, it does not possess any form of plate tectonics, it has a lower gravity, and because of its small size, it cooled more rapidly. The complex geomorphology of the lunar surface has been formed by a combination of processes, chief among which are impact cratering, and volcanism. The Moon is a differentiated body, which possesses a crust, mantle and core.
Geological studies of the Moon are based on a combination of Earth based telescope observations, measurements from orbiting spacecraft, lunar samples, and geophysical data. A few locations were sampled directly during the Apollo missions in the late 1960s and early 1970s, which returned approximately 385 kilograms of lunar rock and soil to Earth, as well as several missions of the Soviet Luna programme. The Moon is the only extraterrestrial body for which we possess samples with a known geologic context. A handful of lunar meteorites have been recognized on Earth, though their source craters on the Moon are unknown. A substantial portion of the lunar surface has not been explored and a number of geological questions remain unanswered.
For a long time, the fundamental question regarding the history of the Moon was of its origin. Early hypotheses included fission from the Earth, capture, and co-accretion. Today, the giant impact hypothesis is widely accepted by the scientific community.
The idea that the early Earth, with an accelerated rotation, expelled a piece of its mass was proposed by George Darwin (son of the famous biologist Charles Darwin). It was commonly assumed that the Pacific Ocean represented the scar of this event. However, today it is known that the oceanic crust that makes up this ocean basin is relatively young, about 200 million years old or less, whereas the Moon is much older. This hypothesis can not account for the angular momentum of the Earth-Moon system.
 Lunar capture
This hypothesis states that the Moon was captured, completely formed, by the gravitational field of the Earth. This is unlikely, since a close encounter with the Earth would have produced either a collision or an alteration of the trajectory of the body in question, so if it had indeed happened, the Moon probably would never return to meet again with the Earth. For this hypothesis to function, there would have to be a large atmosphere extended around the primitive Earth, which would be able to slow the movement of the Moon before it could escape. This hypothesis is considered to explain the irregular satellite orbits of Jupiter and Saturn; nevertheless, it is very difficult to believe that this would explain the origin of our moon. In addition, this hypothesis has difficulty explaining the similar oxygen isotope ratio of the two worlds.
This hypothesis states that the Earth and the Moon formed together as a double system from the primoridial accretion disk of the Solar System. The problem with this hypothesis is that it does not explain the angular momentum of the Earth-Moon system, nor why the Moon is depleted in metallic iron.
Giant impact theory
At present the best explanation for the origin of the Moon involves a collision of two protoplanetary bodies during the early accretional period of Solar system evolution. This "giant impact theory", which became popular in 1984 (although it originated in the mid-1970s) satisfies the orbital conditions of the Earth and Moon and can account for the relatively small metallic core of the Moon. Collisions between planetesimals are now recognized to lead to the growth of planetary bodies early in the evolution of the solar system, and in this framework it is inevitable that large impacts will sometimes occur when the planets are nearly formed.
The theory requires a collision between a body about 90% the present size of the Earth, and another the diameter of Mars (half of the terrestrial radius and a tenth of its mass). The colliding body has sometimes been referred to as Theia, the mother of Selene, the Moon goddess in Greek mythology. This size ratio is needed in order for the resulting system to possess sufficient angular momentum to match the current orbital configuration. Such an impact would have put enough material into orbit about the Earth to have eventually accumulated to form the Moon.
Computer simulations of this event appear to show that the collision must occur with a somewhat glancing blow. This will cause a small portion of the colliding body to form a long arm of material that will then shear off. The asymmetrical shape of the Earth following the collision then causes this material to settle into an orbit around the main mass. The energy involved in this collision is impressive: trillions of tons of material would have been vaporized and melted. In parts of the Earth the temperature would have risen to 10,000 °C.
This formation theory helps explain why the Moon possesses only a small iron core (roughly 25% of its radius, in comparison to about 50% for the Earth). Most of the iron core from the impacting body is predicted to have accreted to the core of the Earth. The lack of volatiles in the lunar samples is also in part explained by the energy of the collision. The energy liberated during the reaccreation of material in orbit about the Earth would have been sufficient to melt a large portion of Moon, leading to the generation of a magma ocean.
The newly formed moon orbited at about one-tenth the distance that it does today, and became tidally-locked with the Earth, where one side continually faces toward the Earth. The geology of the Moon has since been independent of the Earth. While this theory explains many aspects of the Earth-Moon system, there are still a few unresolved problems facing this theory, such as the Moon's volatile elements not being as depleted as expected from such an energetic impact.
The geological history of the Moon has been defined into six major epochs, called the lunar geologic timescale. Starting about 4.5 billion years ago, the newly formed Moon was in a molten state and was orbiting much closer to the Earth. The resulting tidal forces deformed the molten body into an ellipsoid, with the major axis pointed towards Earth.
The first important event in the geologic evolution of the Moon was the crystallization of near global magma ocean. It is not known with certainty what its depth was, but several studies imply a depth of about 500 km, or greater. The first minerals to form in this ocean were the iron and magnesium silicates olivine and pyroxene. Because these minerals were denser than the molten material around them, they sank. After crystallization was about 75% complete, less dense anorthositic plagioclase feldspar crystallized and floated, forming an anorthositic crust about 50 km in thickness. The majority of the magma ocean crystallized quickly (within about 100 million years or less), though the final remaining KREEP-rich magmas, which are highly enriched in incompatible and heat producing elements, could have remained partially molten for several hundred million (or perhaps 1 billion) years. It appears that the final KREEP-rich magmas of the magma ocean eventually became concentrated within the region of Oceanus Procellarum and the Imbrium basin, a unique geologic province that is now known as the Procellarum KREEP Terrane.
Quickly after the lunar crust formed, or even as it was forming, different types of magmas that would give rise to the Mg-suite norites and troctolites began to form, although the exact depths at which this occurred are not known precisely. Recent theories suggest that Mg-suite plutonism was largely confined to the region of the Procellarum KREEP Terrane, and that these magmas are genetically related to KREEP in some manner, though their origin is still highly debated in the scientific community. The oldest of the Mg-suite rocks have crystallization ages of about 3.85 Ga. However, it should be noted that the last large impact that could have excavated deep into the crust (the Imbrium basin) also occurred at 3.85 Ga. Thus, it seems probable that Mg-suite plutonic activity continued for a much longer time, and that younger plutonic rocks exist deep below the surface.
Analysis of the lunar samples seem to imply that a significant percentage of the lunar impact basins formed within a very short period of time between about 4 and 3.85 Ga. This hypothesis is referred to as the lunar cataclysm or late heavy bombardment. However, it is now recognized that ejecta from the Imbrium impact basin (one of the youngest large impact basins on the Moon) should be found at all of the Apollo landing sites. It is thus possible that ages for some impact basins (in particular Mare Nectaris) could have been mistakenly assigned the same age as Imbrium.
The lunar mare represent ancient flood basaltic eruptions. In comparison to terrestrial lavas, these contain higher iron abundances, have low viscosities, and some contain highly elevated abundances of the titanium-rich mineral ilmenite. The majority of basaltic eruptions occurred between about 3 and 3.5 Ga, though some mare samples have ages as old as 4.2 Ga, and the youngest (based on the method of crater counting) are believed to have erupted only 1 billion years ago. Along with mare volcanism came pyroclastic eruptions, which launched molten basaltic materials hundreds of kilometers away from the volcano. A large portion of the mare formed, or flowed into, the low elevations associated with the nearside impact basins. However, it must be noted that Oceanus Procellarum does not correspond to any known impact structure, and the lowest elevations of the Moon within the farside South Pole-Aitken basin are only modesty covered by mare (see lunar mare for a more detailed discussion).
Impacts by meteorites and comets are the only abrupt geologic force acting on the Moon today, though the variation of Earth tides on the scale of the Lunar anomalistic month causes small variations in stresses. (Yu. V. Barkin, J. M. Ferrándiz and Juan F. Navarro, 'Terrestrial tidal variations in the selenopotential coefficients,' Astronomical and Astrophysical Transactions, Volume 24, Number 3 / June 2005, pp. 215 - 236.) Some of the most important craters used in lunar stratigraphy formed in this recent epoch. For example, the Copernicus crater, which has a depth of 3.76 km and a radius of 93 km, is believed to have formed about 900 million years ago (though this is debatable). The Apollo 17 mission landed in an area in which the material coming from the Tycho crater might have been sampled. The study of these rocks seem to indicate that this crater could have formed 100 million years ago, though this is debatable as well. The surface has also experienced space weathering due to high energy particles, solar wind implantation, and micrometeorite impacts. This process causes the ray systems associated with young craters to darken until it matches the albedo of the surrounding surface. However, if the composition of the ray is different from the underlying crustal materials (as might occur when a "highland" ray is emplaced on the mare), the ray could be visible for much longer times.
The lunar landscape is characterized by impact craters, their ejecta, a few volcanoes, hills, lava flows and depressions filled by magma.
The most distinctive aspect of the Moon is the contrast between its light and dark zones. Lighter surfaces are the lunar highlands, which receive the name of terrae (singular terra, from the Latin for Earth), and the darker plains are called maria (singular mare, from the Latin for sea), after Johannes Kepler who introduced the name in the 1600s. The highlands are anorthositic in composition, whereas the maria are basaltic. The maria often coincide with the "lowlands," but it is important to note that the lowlands (such as within the South Pole-Aitken basin) are not always covered by maria. The highlands are older than the visible maria, and hence are more heavily cratered.
The major products of volcanic processes on the Moon are evident to the Earth-bound observer in the form of the lunar maria. These are large flows of basaltic lava that correspond to low-albedo surfaces covering nearly a third of the near side. Only a few percent of the farside has been affected by mare volcanism. Even before the Apollo missions confirmed it, most scientists believed that the maria were lava-filled plains, since they possessed lava flow patterns and collapses attributed to lava tubes.
The ages of the mare basalts have been determined both by direct radiometric dating and by the technique of crater counting. The oldest radiometric ages are about 4.2 Ga, whereas the youngest ages determined from crater counting are about 1 Ga (1 Ga = 1 billion years old). Volumetrically, most of the mare formed between about 3 and 3.5 Ga. The youngest lavas erupted within Oceanus Procellarum, whereas some of the oldest appear to be located on the farside. The maria are clearly younger than the surrounding highlands given their lower density of impact craters.
A large portion of maria erupted within, or flowed into, the low lying impact basins on the lunar nearside. Nevertheless, it is unlikely that a causal relationship exists between the impact event and mare volcanism because the impact basins are much older (by about 500 million years) than the mare fill. Furthermore, Oceanus Procellarum, which is the largest expanse of mare volcanism on the Moon, does not correspond to any known impact basin. It is commonly suggested that the reason the mare only erupted on the nearside is that the nearside crust is thinner than the farside. While crustal thickness variations might act to modulate the amount of magma that ultimately reaches the surface, this hypothesis does not explain why the farside South Pole-Aitken basin, whose crust is thinner than Oceanus Procellarum, was only modestly filled by volcanic products. Finally, it should be noted that the Earth's gravity played no preferential role in causing mare volcanism to occur on the near side, as the Earth's gravitational attraction is exactly balanced by the centrifugal acceleration resulting from the Moon's rotation.
Another type of deposit associated with the mare, although it also covers the highland areas, are the "dark mantle" deposits. These deposits cannot be seen with the naked eye, but they can be seen in images taken from telescopes or orbiting spacecraft. Before the Apollo missions, scientists believed that they were deposits produced by pyroclastic eruptions. Some deposits appear to be associated with dark elongated ash cones, reinforcing the idea of pyroclasts. The existence of pyroclastic eruptions was later confirmed by the discovery of glass spherules similar to those found in pyroclastic eruptions here on Earth.
Many of the lunar basalts contain small holes called vesicles, which were formed by gas bubbles exsolving from the magma at the vacuum conditions encountered at the surface. It is not known with certainty which gases escaped these rocks, but carbon monoxide is one candidate.
The samples of pyroclastic glasses are of green, yellow, and red tints. The difference in color indicates the concentration of titanium that the rock possesses, with the green particles having the lowest concentrations (about 1%), and red particles having the highest concentrations (up to 14%, much more than the basalts with the highest concentrations).
Basaltic eruptions on the Moon sometimes resulted in the formation of localized lava channels. Rilles generally fall into three categories, consisting of sinuous, arcuate, or linear shapes. By following these meandering rilles back to their source, they often lead to an old volcanic vent. One of the most notable sinuous rilles is the Vallis Schröteri feature, located in the Aristarchus plateau along the eastern edge of Oceanus Procellarum.
A variety of shield volcanos can be found in selected locations on the lunar surface, such as on Mons Rümker. These are believed to be formed by relatively viscous, possibly silica-rich lava, erupting from localized vents. The resulting lunar domes are wide, rounded, circular features with a gentle slope rising in elevation a few hundred meters to the mid-point. They are typically 8-12 km in diameter, but can be up to 20 km across. Some of the domes contain a small pit at their peak.
Wrinkle ridges are features created by compressive tectonic forces within the mare. These features represent buckling of the surface and form long ridges across parts of the mare. Some of these ridges may outline buried craters or other features beneath the mares. A prime example of such an outlined feature is the Letronne crater.
Graben are tectonic features that form under extension stresses. Structurally, they are composed of two normal faults, with a down-dropped block between them. Most graben are found within the lunar mare near the edges of large impact basins.
Mare Imbrium and the Copernicus crater. NASA photo.
It may be surprising to learn that the origin of the Moon's craters as impact features became widely accepted only in the 1940s. This realization allowed the impact history of the Moon to be gradually worked out by means of the geologic principle of superposition. That is, if a crater (or its ejecta) overlaid another, it must be the younger. The amount of erosion experienced by a crater was another clue to its age, though this is more subjective. Adopting this approach in the late 1950s, Gene Shoemaker took the systematic study of the Moon away from the astronomers and placed it firmly in the hands of the lunar geologists.
Impact cratering is the most notable geological process on the Moon. The craters are formed when a solid body, such as an asteroid or comet, collides with the surface at a high velocity (mean impact velocities for the Moon are about 17 km per second). The kinetic energy of the impact creates a compression shock wave that radiates away from the point of entry. This is succeeded by a rarefaction wave, which is responsible for propelling most of the ejecta out of the crater. Finally there is a hydrodynamic rebound of the floor that can create a central peak.
These craters appear in a continuum of diameters across the surface of the Moon, ranging in size from tiny pits to the immense South Pole-Aitken Basin with a diameter of nearly 2,500 km and a depth of 13 km. In a very general sense, the lunar history of impact cratering follows a trend of decreasing crater size with time. In particular, the largest impact basins were formed during the early periods, and these were successively overlaid by smaller craters. The size frequency distribution (SFD) of crater diameters on a given surface (that is, the number of craters as a function of diameter) approximately follows a power law with increasing number of craters with decreasing crater size. The vertical position of this curve can be used to estimate the age of the surface.
King crater displays the characteristic features of a large impact formation, with a raised rim, slumped edges, terraced inner walls, a relatively flat floor with some hills, and a central ridge. The Y-shaped central ridge is unusually complex in form. NASA photo.
The most recent impacts are distinguished by well-defined features, including a sharp-edged rim. Small craters tend to form a bowl shape, while larger impacts can have a central peak with flat floors. Larger craters generally display slumping features along the inner walls that can form terraces and ledges. The largest impact basins, the multiring basins, can even have secondary concentric rings of raised material.
The impact process excavates high albedo materials that initially gives the crater, ejecta, and ray system a bright appearance. The process of space weathering gradually decreases the albedo of this material such that the rays fade with time. Gradually the crater and its ejecta undergo impact erosion from micrometeorites and smaller impacts. This erosional process softens and rounds the features of the crater. The crater can also be covered in ejecta from other impacts, which can submerge features and even bury the central peak.
The ejecta from large impacts can include larges blocks of material that reimpact the surface to form secondary impact craters. These craters are sometimes formed in clearly discernible radial patterns, and generally have shallower depths than primary craters of the same size. In some cases an entire line of these blocks can impact to form a valley. These are distinguished from catena, or crater chains, which are linear strings of craters that are formed when the impact body breaks up prior to impact.
Generally speaking a lunar crater is roughly circular in form. It has been demonstrated in laboratory experiments that even low-angle impacts tend to produce circular craters, and there are few lunar craters that have formed naturally elliptical outlines. However, a low angle impact can produce a central peak that is offset from the mid-point of the crater. In addition, such an impact will produce an asymmetrical ejecta and ray system.
Dark-halo craters are formed when an impact excavates lower albedo material from beneath the surface, then deposits this darker ejecta around the main crater. This can occur when an area of darker basaltic material, such as that found on the lunar mare, is later covered by lighter ejecta derived from more distant impacts in the highlands. This covering conceals the darker material below, which is later excavated by subsequent craters.
The largest impacts produced melt sheets of molten rock that covered portions of the surface which could be as thick as a kilometer. Examples of such impact melt can be seen in the northeastern part of the Mare Orientale impact basin.
The surface of the Moon has been subject to billions of years of collisions with both small and large asteroidal and cometary materials. Over time, these impact processes have pulverized and "gardened" the surface materials, forming a fine grained layer termed "regolith". The thickness of the regolith varies between 2 meters beneath the younger maria, to up to 20 meters beneath the oldest surfaces of the lunar highlands. The regolith is predominantly composed of materials found in the region, but also contains traces of materials ejected by distant impact craters. The term "mega-regolith" is often used to describe the heavily fractured bedrock directly beneath the near-surface regolith layer.
The regolith contains rocks, fragments of minerals from the original bedrock, and glassy particles formed during the impacts. In most of the lunar regolith, half of the particles are made of mineral fragments fused by the glassy particles; these objects are called agglutinates. The chemical composition of the regolith varies according to its location; the regolith in the highlands is rich in aluminium and silica, just as the rocks in those regions. The regolith in the maria is rich in iron and magnesium and is silica-poor, as the basaltic rocks from which it is formed.
The lunar regolith is very important because it also stores information about the history of the Sun. The atoms that compose the solar wind – mostly helium, neon, carbon and nitrogen – hit the lunar surface and insert themselves into the mineral grains. Upon analyzing the composition of the regolith, particularly its isotopic composition, it is possible to determine if the activity of the Sun has changed with time. The gases of the solar wind could be useful for future lunar bases, since oxygen, hydrogen (water), carbon and nitrogen are not only essential to sustain life, but are also potentially very useful in the production of fuel. The composition of the lunar regolith can also be used to infer its source origin.
The first rocks brought back by Apollo 11 were basalts. In spite that the mission landed on Mare Tranquillitatis, a few millimetric fragments of rocks coming from the highlands were picked up. These are composed mainly of plagioclase feldspar; some fragments were composed exclusively of anorthositic plagioclase. The identification of these mineral fragments led to the bold hypothesis that a large portion of the Moon was once molten, and that the crust formed by fractional crystallization of this magma ocean.
A natural outcome of the giant impact event is that the materials that reaccreted to form the Moon must have been hot. Current models predict that a large portion of the Moon would have been molten shortly after the Moon formed, with estimates for the depth of this magma ocean ranging from about 500 km to full moon melting. Crystallization of this magma ocean would have given rise to a differentiated body with a compositionally distinct crust and mantle and accounts for the major suites of lunar rocks.
As crystallization of the lunar magma ocean proceeded, minerals such as olivine and pyroxene would have precipitated and sank to form the lunar mantle. After crystallization was about three-quarters complete, anorthositic plagioclase would have begun to crystallize, and because of its low density, float, forming an anorthositic crust. Importantly, elements that are incompatible (i.e., those that partition preferentially into the liquid phase) would have been progressively concentrated into the magma as crystallization progressed, forming a KREEP-rich magma that initially should have been sandwiched between the crust and mantle. Evidence for this scenario comes from the highly anorthositc composition of the lunar highland crust, as well as the existence of KREEP-rich materials.
The Apollo program brought back 381.7 kg (841.5 lb) of lunar surface material, most of which is stored at the Lunar Receiving Laboratory in Houston, Texas. These rocks have proved to be invaluable in deciphering the geologic evolution of the Moon. Lunar rocks are in large part made of the same common rock forming minerals as found on Earth, such as Olivine, Pyroxene, and Plagioclase Feldspar (Anorthosite). The mineral ilmenite is highly abundant in some mare basalts, and a new mineral named Armalcolite (named for Armstrong, Aldrin, and Collins, the three members of the Apollo 11 crew) was first discovered in the lunar samples.
The maria are composed predominantly of basalt, whereas the highland regions are iron-poor and composed primarily of anorthosite, a rock composed primarily of calcium-rich plagioclase feldspar. Another significant component of the crust are the igneous Mg-suite rocks, such as the troctolites, norites, and KREEP-basalts. These rocks are believed to be genetically related to the petrogenesis of KREEP.
Composite rocks on the lunar surface often appear in the form of breccias. Of these, the subcategories are called fragmental, granulitic, and impact-melt breccias, depending on how they were formed. The mafic impact melt breccias, which are typified by the low-K Fra Mauro composition, have a higher proportion of iron and magnesium than typical upper crust anorthositic rocks, as well as higher abundances of KREEP.
Composition of the maria
The main characteristics of the basaltic rocks with respect to the rocks of the lunar highlands is that the basalts contain higher abundances of olivine and pyroxene, and less plagioclase. They are more rich in iron than terrestrial basalts, and also have lower viscosities. Some of them have high abundances of a ferro-titanic oxide called ilmenite. Since the first sampling of rocks contained a high content of ilmenite and other related minerals, they received the name of "high titanium" basalts. The Apollo 12 mission returned to Earth with basalts of lower titanium concentrations, and these were dubbed "low titanium" basalts. Subsequent missions, including the Soviet unmanned probes, returned with basalts with even lower concentrations, now called "very low titanium" basalts. The Clementine space probe returned data showing that the mare basalts possess a continuum in titanium concentrations, with the highest concentration rocks being the least abundant.
The current model of the interior of the Moon was derived using seismometers left behind during the manned Apollo program missions, as well as investigations of the Moon's gravity field and rotation.
The mass of the Moon is sufficient to eliminate any voids within the interior, so it is believed to be composed of solid rock throughout. Its low bulk density (~3346 kg m-3) indicates a low metal abundance. Mass and moment of inertia constraints indicate that the Moon likely has an iron core that is less than about 450 km in radius. Studies of the Moon's physical librations (small perturbations to its rotation) furthermore indicate that the core is still molten. Most planetary bodies and moons have iron cores that are about half the size of the body. The Moon is thus anomalous in possessing a core whose size is only about one quarter of its radius.
The crust of the Moon is on average about 50 km thick (though this is uncertain by about ±15 km). It is widely believed that the far-side crust is on average thicker than the near side by about 15 km. Seismology has constrained the thickness of the crust only near the Apollo 12 and 14 landing sites. While the initial Apollo-era analyses suggested a crustal thickness of about 60 km at this site, recent reanalyses of this data set suggest a thinner value, somewhere between about 30 and 45 km.
Compared to that of Earth, the Moon has only a very weak external magnetic field. Other major differences are that the Moon does not currently have a dipolar magnetic field (as would be generated by a geodynamo in its core), and the magnetizations that are present are almost entirely crustal in orgin. One hypothesis holds that the crustal magnetizations were acquired early in lunar history when a geodynamo was still operating. The small size of the lunar core, however, is a potential obstacle to this theory. Alternatively, it is possible that on airless bodies such as the Moon, transient magnetic fields could be generated during impact processes. In support of this, it has been noted that the largest crustal magnetizations appear to be located near the antipodes of the largest impact basins. While the Moon does not possess a dipolar magnetic field like the Earth does, some of the returned rocks possess strong magnetizations. Furthermore measurements from orbit show that some portions of the lunar surface are associated with strong magnetic fields.