On any good map or photograph of the Moon one finds several extended smooth areas of a darker color than the rest of the lunar surface. These are the so-called maria, or seas, which are believed to be extensions of solidified lava. The rocks brought back by the Apollo 11 astronauts lend support to this interpretation of the nature of lunar maria. Several of the maria—among them some of the most prominent are Crisium, Serenitatis and Imbrium—have a rounded shape strongly reminiscent of impact areas by sizable celestial bodies. Smaller impact areas which present the same smooth and darkly colored surface are the craters Plato, Archimedes, Ptolemaeus, Billy, Crtiger, Grimaldi and several others.
The origin of the lava, which filled these presumed impact areas, has for a long time been a matter of scientific dispute. One of the oldest theories, which until recently has been finding wide support, holds that the lava is of volcanic, or internal, origin. Another theory, which has been prominent since Gilbert presented it in 1893, ascribes the lava formation to heat generated by the impacts in which the respective maria and craters were created. For the production of large quantities of lava the velocity of the colliding object would have to be within a narrow range of 2 to 3 km/sec. A question to be clarified is whether and how this kind of velocity is possible in a lunar impact.
The way in which some of the lunar surface features were formed appears easier to understand if one distinguishes between satellite impacts and meteor impacts. The former are due to collisions involving objects in orbit around the Earth, while the latter are due to collisions involving objects which are not Earth satellites, such as meteors and comets. The velocity and energy generated in a satellite impact has long been known to be primarily created through the acceleration of the impact object by the Moon's gravitational field, a fact repeatedly verified also by data-machine trials. This means that the collision velocity of a satellite impact is usually close to the lunar escape velocity of 2.4 km./sec., which is well within the range of 2 to 3 km./sec. mentioned above in which large quantities of lava can be produced. Off the other hand, the meteor impact velocity is often in the same range as the Earth's and Moon's orbital velocity in their movement around the Sun, or ca. 30 km/sec. This is one order of magnitude greater than the lunar escape velocity. The energies and temperatures generated are proportional to the square of the impact velocity. Temperatures above 100,000 degrees Kelvin can easily arise in a meteor impact and the energy generated during such impacts is usually sufficient to vaporize a substantial portion of the material involved in the collision, causing an explosion and creating a saucer-shaped crater similar to the large meteor craters on Earth—Marshall, 1943. On the other hand, although the energy and temperatures generated during a satellite impact are usually sufficient to smelt part of the material involved in the collision, they are insufficient to vaporize a significant quantity of the material.
Thus, lunar surface structures caused by satellite impacts tend to present a dark, smooth, horizontal surface indicating the production of large amounts of lava, while those caused by the meteor impacts give evidence of a violent explosion in which most of the material was vaporized, leaving behind comparatively much less lava. An effect of the explosion is to cancel any possible trace of the inclination with which the meteor reached the lunar surface. Meteor craters, therefore, will have the same saucer-shape appearance irrespective of the impacting angle. Not necessarily so in the satellite impact areas, which occasionally may retain traces of the inclination and direction from which the impacting object reaches the lunar surface. This is particularly evident in the case of the Imbrium impact which will be our main interest.
Following these criteria the impact maria and those craters, which present the same smooth darkly colored surface characteristic of maria, have been ascribed to collisions with smaller satellites that may have been in orbit around the Earth in the early days of our satellite system. On the other hand, the more typical saucer-shaped lunar craters are ascribed to the violent explosions which would result from collisions with the much faster objects, like meteors and small planets, that were not part of the Earth's satellite system.
We shall present here one of the main points of major interest in the impact theory, namely the interpretation of several prominent features and characteristics of the Mare Imbrium and surrounding areas.
The Mare Imbrium Region
The Mare Imbrium is the largest of the circular maria. It is easily visible in the northern—upper in Europe and North America—hemisphere, a little to the left of the central meridian, on a binocular or even by naked eye. On a good map or photograph (see Page 68) one can notice two distinct areas inside this mare: (1) A central roughly circular area which is very smooth except for a few comparatively small craters due to late impacts which occurred after the lava was solidified. (2) A horseshoe shaped border area nearly surrounding the central part from all sides, except the northwestern sector, with much greater crater density and containing several mountains and large craters notably Archimedes, Aristillus, the Lower Apennine Highlands, the Teneriffe Mountains, and several others.
The smooth central area has a diameter of roughly 700 km., while the diameter of the whole Mare Imbrium is roughly 1,100 km. The size of the mare, the asymmetric shape of the horseshoe border area and the other characteristics of the whole region which will be described below, have led to the following interpretation of the events which may have shaped this remarkable feature of the lunar surface. Probably between 3.5 and 4 billion years ago one of the early satellites of Earth—which we may call "The Imbrium Satellite"—collided with the Moon after approaching its surface from a low angle to the horizon. The main body of the satellite approached the lunar surface moving from a northwesterly toward a southeasterly direction and reached the Moon either in the Sinus Iridium—according to Urey—or a couple hundred kilometers farther to the southeast, nearer the center of the smooth circular area—according to Gilbert's and the author's interpretations.
The Impact
The Imbrium satellite must have been the largest—or one of the largest—satellite that ever collided with the Moon. Its diameter may have been as large as 300 to 350 km. Before reaching the lunar surface several fragments may have been separated from the main body of the satellite by the gravitational pull of the Moon, after the satellite entered the Roche limit.* (*The Roche limit is a distance from the center of a planet, or satellite corresponding to approximately 2.3 planetary or satellite radii. Objects within this distance tend to be broken up by the gravitational field of the planet or satellite into smaller fragments. This concept is used among others in the theory of the formation of the rings around Saturn.) Many more smaller fragments may have been formed after the satellite strafed the lunar surface in its nearly horizontal path. The impact of this satellite must have been a spectacular show.
First, a major fragment of the satellite may have reached the Sinus Iridium producing a crater of about 210 km. in diameter. After one or two minutes the main body of the satellite—or its main fragments—may have reached the lunar surface one or two hundred kilometers farther to the southeast, between the Sinus Iridium and the center of the circular smooth area already described. The temperature produced by the impact was not by far as high as the temperature—often several hundred thousands of degrees centigrade—produced by major meteors. This is because the velocity of an Earth satellite colliding with the Moon is unlikely to be much higher than 2.4 km./sec.—the Moon's escape velocity. Probably the highest temperatures did not exceed 2,000 degrees centigrade. But the phenomenon was nonetheless most spectacular.
Within minutes a roughly circular area with a diameter of about 700 km. was smelted. The shock wave produced by the impact had also demolished almost completely the surrounding horseshoe border area of the Mare Imbrium and was proceeding to farther distances generating what may probably have been the most gigantic Moon-quake in history. After the shock wave several kilometers high waves of lava originated in the molten part of the Mare Imbrium and completed the destruction of the horseshoe border area. A portion of the lava may have crossed the mountains to the north and flooded the regions behind, thus forming the Mare Frigoris and the Sinus Roris. A portion of the Oceanus Procellarum may also have been flooded on this occasion. At the same time a gigantic shower, of various sized fragments of the Imbrium satellite, was reaching the lunar surface. The main stream of these fragments was moving, like the Imbrium satellite from northwesterly to southeasterly direction.
A large region to the southeast of the Mare Imbrium, and particularly the Haemus mountains, shows clear signs of having been deformed and reshaped by the gigantic stream of solid fragments, as can be seen on Page 68. The whole region presents the most startling testimony of the gigantic storm which followed the Imbrium impact as pointed out by Gilbert, Urey and others. Some of the largest fragments produced the craters in the horseshoe border areas which have already been mentioned, and many of the craters which have apparently the same age and are spread in a large region to the southeast of the Imbrium and the Haemus mountains, such as the craters Conou, Manilius. Menelaus, Plinius, Ross and Arago. Some of the fragments may have reached the horseshoe border area in the brief interval between the shock wave and the following waves of molten lava, producing ghost craters like Wallace and possibly Cassini (see Page 68).
Lunar Spin After the Impact
Most of the facts presented above have been known since Gilbert presented the satellite impact theory in 1893. Nevertheless the fact, that a satellite the size of Imbrium reaching the lunar surface on a nearly horizontal path at a velocity of roughly 2.4 km./sec. would almost certainly have left the Moon spinning with a significant velocity, is one which has long escaped our attention. The mass of the Imbrium satellite is estimated to have been at least 1/1000 of the Moon's own mass. If the Moon had been at rest at the time of the impact, its rotation period after the impact would have been less than thirty days. This amount of spin would be quite sufficient to bring the Moon out of its bound rotation to the Earth today, and would almost certainly have been sufficient to do the same at the time the collision occurred—see Barricelli and Metcalf 1969. Only after the problem of explaining the distribution of maria on the lunar surface, and particularly their nearly total absence on the far side of the Moon arose as a major issue, attention was focused on the possibility that the face of the Moon, which today faces the Earth, at one time may have been the far side of the Moon.
Together with R. Metcalf, the author of this article had undertaken the task of investigating whether the asymmetric distribution of maria could be explained by the following facts:
1. The Moon's distance from the Earth is known to have been increasing steadily as a result of the tides effect. Between 4 and 4.5 billion years ago the Moon may have been very close to the Earth's surface and perhaps barely outside of the Roche limit.
2. Under these conditions the Moon could only collide with satellites in orbits external to the Moon itself since it is well-known that any sizable satellite inside the Roche limit would rapidly disintegrate into some kind of saturnic ring.
This theory can explain the fact that the Earth has no other satellites today, since earlier satellites would either have collided with or have been deviated by the Moon as its distance from the Earth increased. Even if the Earth had some kind of saturnic rings, they would have been perturbated and demolished as the various fragments in each ring successively came into resonance with the receding Moon. We undertook the task of investigating by a data processing machine the orbits of a large number of external satellites colliding with the Moon under various conditions. The purpose of our investigation was to find out whether collisions with satellites in external orbits would present an asymmetric distribution on the lunar surface matching the observed distribution of maria. Our findings showed that it would and that the match would be almost perfect if the side of the Moon presently facing the Earth had been the far side of the Moon when the impacts, which created the maria, occurred.
This brought our attention to the fact that collisions, and particularly a collision with a satellite following a nearly horizontal path like the Imbrium collision, might spin the Moon around. Our attention was focused on the Imbrium which is also the largest single impact that can be seen on the Moon and—as pointed out by Urey—is probably also the latest major satellite impact—possibly with the exception of the Mare Orientale whose nature satellite, or meteor impact, is not established with certainty. This led us to recognize the various conditions indicating that the Moon may have been left spinning at an appreciable speed after the impact. None of the other maria present such good evidence that the respective collisions would have disengaged the Moon from its Earthbound rotation.
Earth's Early Satellite System
Our present interpretation of the events leading to the formation of the maria can be summarized as follows:
In the early days of the satellite system, probably around 4.5 billion years ago, several moons of different sizes rotated in the Earth's equatorial plane. The innermost was the largest, and its distance from the Earth was gradually increasing. At a greater distance several other moons, among them the three which originated the maria: Crisium, Serenitatis and Imbrium, rotated in different planes with different inclinations. As the largest moon's distance from the Earth gradually increased it collided first with several of the moons in the equatorial plane. These bodies must have reached the surface of what at that time was the far side of the Moon near the equator, thus creating most equatorial maria like Tranquillitatis, Foecunditatis, and many others. The most recent impacts such as Serenetatis and Imbrium occurred at a greater distance from the equator because of the greater inclinations of external satellites in our system as in all other major satellite systems. The last major collision was the Imbrium impact which left the Moon spinning probably for several thousands, or even millions, of years. When the Moon's Earth-bound rotation was eventually reestablished, the original far side of the Moon and the side originally facing the Earth had been interchanged. This interchanging was made possible by the ellipsoidal symmetry of the Moon which allows two equally stable equilibrium configurations.
This is, according to the impact theory as far as we can tell today, the most likely succession of events which may have led to the present shape and distribution of maria on the lunar surface.
Impact Generated Temporary Atmosphere
The heat generated in a major impact, such as the Imbrium impact, might cause the formation of considerable amounts of carbon dioxide and other gases escaping from lava and overheated material, thus temporarily supplying the Moon with a thin atmosphere. Within a few thousand years most of the atmosphere may have escaped from the Moon, whose gravitational field is incapable of retaining the main atmospheric gases for a long time. A minor portion of this atmosphere may, however, still be on the Moon in a frozen condition in the southern and northern polar regions on the bottom of deep valleys and craters which never are reached by the sun. Temperatures approaching absolute zero may prevail in those places where solar heating is never available to replace the energy lost by irradiation into space. Any kind of common atmospheric gases could remain in these places in a frozen condition for an indefinite length of time. Not only carbon dioxide, but possibly even hydrogen compounds like water, ammonia and methane might be found in a frozen condition in a few places around the polar regions. The solar wind and impacts by comets might have supplied the hydrogen.
Deposits of frozen gases might prove highly valuable in the future, not only for the supply of oxygen and water for human settlements, but also as a source of rocket fuels. The transportation of spacecraft fuels from Earth is extremely expensive, and it is quite conceivable that the first mining industry on the Moon might be aimed at the production of fuels for interplanetary travel. But apart from this possibility an exploration of one of the oldest deep craters in a polar region of the Moon would, in its own right, be worth considering as a possible target—for an Apollo expedition for its purely scientific interest. Few things could bring more light on the past history of the Moon than the discovery of frozen gases deposited at various times since the Moon was formed. Unfortunately old deposits of frozen gases might be covered by a deep layer of dust from meteoric impacts and inaccessible to astronauts with their present equipment. However, nobody knows what there is to be found before some of the coolest spots on the lunar surface have been visited. One certainly can not take it for granted that the surface composition in such places would not be quite different from the rest of the Moon.