COSMIC

HISTORY

INFLUENCES

ON THE EARLY

OF THE LUNAR

SURFACE*

ZDENI~K K O P A L The Lunar Science Institute, Houston, Texas, U.S.A. **

(Received 8 December, 1971) Abstract. It is pointed out that the observed random distribution of low-angle impact craters over the lunar surface rules out the possibility that particles initially responsible for the origin of such craters had, prior to impact, been in heliocentric orbits. The observed facts are more consistent with a view that particles responsible for most of large primary impact at the earliest stage of lunar history were moving with the Earth-Moon gravitational dipole, and may have represented leftovers from the formation of this pair of cosmic bodies. The application of a similar argument to an equally obvious lack of directional effects in Martian cratering is, however, invalidated by a relatively large inclination of the Martian equator to the orbital plane of this planet. As is well known, the stony relief o f the lunar face represents a cumulative record of events - both external and internal - which disfigured the M o o n since the time of its formation; and recent results of work on the chronology o f lunar rocks have demonstrated most part o f this relief to be of very ancient date. As far as the continental areas on the M o o n are c o n c e r n e d - a n d these constitute more than 80% o f the entire lunar s u r f a c e - the formations still extant were accreted probably in the first few hundred million years (possibly less) o f the lunar existence; and the formation of the different maria on the M o o n represented apparently a series of isolated episodes spread out between 3-4 billion years before our time; after which the m o r p h o l o g y o f the surface o f our satellite seems to have undergone few if any important large-scale changes. It will be the purpose o f the present paper to examine, from this point of view, the testimony which the main features of the lunar surface - preserved in their pristine state by a virtual lack of any kind o f erosion - bear on the cosmic environment in which our satellite acquired its principal morphological characteristics. In doing so we shall assume as a working hypothesis that, in the earliest stage of the history o f our satellite, the lunar sculpture in the continental areas was predominantly shaped by external impacts; and that, in particular, the majority of all large craters on the M o o n represent surface scars produced at that time by primary intruders, then more a b u n d a n t in space than they became (on account of depletion) in subsequent ages. Whether or not individual crater formations could also have been produced by other (internal) processes will be irrelevant for our argument as long as the numbers o f such formations do not become statistically significant. On the assumption, therefore, that the majority of old craters in the continental * Presented at the NATO Advanced Study Institute on Lunar Studies in Patras, Greece, September 1971. ** Normally at the Department of Astronomy, University of Manchester, England. The Moon 5 (1972) 200-205. All Rights Reserved Coovrieht © 1972 by D. Reidel Publishine Comlganv, Dordrecht-Holland

Fig. I. An oblique view of the western portion of Mare Foecunditatis, taken on 10 August, 1967, by Orbiter 5 from an altitude of 97 kin. The two craters near the centre of the field are Messier and Messier A, associated with a twin bright ray akin to a cometary tail.

Fig. 2.

Bright rays from Messier in Mare Foecunditatis (above) and Proclus on the Western shores of Mare Crisium (Observatoire du Pic-du-Midi photograph; 24 in. refractor).

202

ZDENEK KOPAL

areas represent surface scars due to primary impacts, their morphological characteristics can tell something about the statistical distribution of directions from which our satellite suffered hits from outside. Extensive laboratory simulations of the cratering processes reported at this conference by Gault (1972) have disclosed that, for angles of impact in excess of (approximately) 15°, the characteristics of the astroblemes produced on the lunar surface can disclose next to nothing as to the direction from which the respective cosmic intruder approached our satellite. For oblique impacts inclined by less than 10°-15 ° to a plane tangent to the lunar surface the direction of the intruder can, however, be inferred from an elongated form of the resultant craters (see Figure 1), or from an unequal distribution ('zone of avoidance') of the apron of the ejecta scooped up by the impact (Figure 2). Impacts oblique enough to preserve the characteristics of the original direction of approach are, to be sure, likely to constitute only a very small fraction of all such events recorded on the lunar face. Nevertheless, the total number of impacts suffered by the Moon in the course of its long astronomical past from all directions is so large that the number of oblique events among them should not be negligible. In point of fact, directional characteristics of low-angle impacts are indicated for a few hundred crater formations exceeding 1 km in size (a few dozen of which are larger than 10 km) on the front side of the Moon alone; and those of this type smaller than 1 km are too many to be individually counted. Such being the size of the sample available for statistical analysis, the question is bound to arise: what is the distribution of low-angle impacts which the Moon has suffered since the time of its formation? The stony sculpture of the continental areas goes back, in its structure, probably all the way to the age of 4.6 billion years established for the bulk of the lunar 'fines' returned by successive Apollo missions of 1969-72; and while large-scale features (in excess of, say, 100 km in size) may have partly fallen prey to the 'saturation bombardment' suffered at the earliest stage of lunar history (cf. Marcus, 1966), this is not likely to be true of the smaller crater formations preserving directional characteristics. Therefore, their cumulative sample available for our analysis should have reached us essentially unimpaired from the earliest days of lunar history, and provide a link with the most distant past of the solar system. The answer to our question turns out to be simple, but very interesting in its cosmogonic implications: namely, the distribution of craters due to oblique impacts appears to be essentially at random all over the lunar surface - and does not exhibit clustering in any particular zone. The full significance of this result will emerge only when we consider it in connection with the orientation of the Moon in space, and the direction if its axis of rotation with respect to the invariable plane of the solar system. As is well known, the inclination I of the lunar equator to the ecliptic is very small only 1°32'4"± 7" according to its latest determination by Koziel (1967a, b). An even more meaningful result will be obtained if we refer this inclination to the invariable plane of the solar system, the present orientation of which with respect to the ecliptic is specified by the angles f~= 106°44 ` and i = 1°38'58 " for the epoch of 1900.0 (cf.

COSMIC INFLUENcEs ON THE EARLY HISTORY OF THE LUNAR SURFACE

203

Clemence and Brouwer, 1955). Koziel's value of I given above, and deduced from the lunar heliometric observations carried out between 1877-1915, should refer approximately to the same epoch; and if so, the inclination of the Moon's axis of rotation (as specified by the angles I = l°32'4" and f2=n) to the invariable plane of the solar system around 1900.00 must have been equal to 119oo=2°1 '. This latter value cannot, to be sure, remain secularly constant; for the longitude of the node of the invariable plane is known to be increasing (at present) by 59' pet" century, and its inclination diminishing by 18" per century (cf. Clemence and Brouwer, 1955) - no doubt as a result of an action of harmonic terms of very long period - and while nothing as yet is known directly about the nutation of the lunar axis of rotation, its amplitude is unlikely to be zero, and its period must be very small a fraction of the total age of our satellite. Therefore, all we can say at present about the inclination of the lunar axis of rotation to the invariable plane of the solar system is that it oscillates between i+_I, or 0°-3°; and may have done so throughout the entire astronomical past of our satellite. But if so, it would follow that its equatorial belt, or polar caps, have never been very far from the positions which they occupy at the present time; and their latitudes may have secularly changed by only a few degrees. Now let us examine, in this light, the fact that evidence for impacts in all directions including low-angle or grazing impacts - appears to be distributed at random all over the surface of our satellite; and inquire about the orbits which the impacting particles must have followed prior to their collisions with the Moon. If these particles had been in heliocentric orbits, the inclinations of such orbits could not have deviated much from the ecliptic, or they would have missed the Moon altogether; for only particles in orbits closely parallel with the ecliptic could have impinged anywhere on the surface of our satellite. But if the Moon's axis of rotation has been almost perpendicular to the ecliptic, it would follow that high-angle impacts should have occurred primarily in the equatorial zones; and low-angle or grazing impacts limited almost exclusively to the polar regions of our satellite. The extant stony record of the lunar face discloses conclusively that this is not the case; but, rather, that impacts at all angles appear to be distributed at random in all parts of the Moon. But if so, it follows that the impinging particles could not have revolved in heliocentric orbits, but must have been moving in trajectories which made impacts at all angles equally likely all over the surface of our satellite. In other words, the particles destined to end up their cosmic careers by collisions with the lunar surface were not moving prior to impacts around the Sun, but within the domain of influence of the Earth-Moon gravitational dipole; with the Sun exerting only perturbations of their motions at a distance. Only in this way could impacts in all directions become equally likely. T h i s - if t r u e - would suggest that the particles responsible for the bulk of the cosmic bombardment of the lunar surface at the earliest stage of its history were those left over in the Earth-Moon neighborhood from the process which led to the formation of this pair of cosmic bodies, and carried along with them through space until most of them were swept up by the gravitational attraction of the two mass centres. This

204

ZDENEK KOPAL

would, in turn, imply that the Earth and the Moon were already gravitational partners (though not necessarily at their present distance) during the first few hundred million years of their existence - an epoch in which the cratering of the lunar surface is now thought to have been largely accomplished - and if the Moon was captured by the Earth, this event must have occurred prior to the main period of cratering of the lunar continental land masses. Is there any escape from these conclusions? One would be to assume that most part of the continental impact craters were inflicted on the M o o n before its eventual capture by the Earth - when the Moon, travelling alone through space, may have been tumbling about its centre of mass, so that impacts from all directions would have been equally likely. Such a view is, however, highly improbable. It is true that the present rate of axial rotation of the Moon, and its physical librations in latitude, longitude, and node are virtually controlled by the Earth (and would have been so even more effectively at a shorter distance between the two bodies). We have, therefore, no idea whether or not the M o o n could have been spin-stabilized in space before its hypothetical capture. However, even if this was not the case, it is most unlikely that any of its pre-existing surface sculpture would have survived so brutal an experience as a capture of the M o o n by the Earth would inevitably have been. The need to slow down a passer-by to make capture possible would have required a dissipation of so much energy of the Moon through inelastic tides as could be raised only at a very close approach; and such tides would no doubt have obliterated most, if not all, of the pre-existing surface markings. Therefore, if the Moon was captured by the Earth at the earliest stage of the history of the solar system, its present surface sculpture must have begun to accumulate after the time of the capture; and if it was formed as an independent body in the proximity of the Earth to begin with, its inclination to the invariable plane of the solar system has probably remained more or less unchanged up to the present time. But, either way, the argument against heliocentric orbits of particles whose collisions with the M o o n disfigured most part of its continental areas continues to hold good. In conclusion, it should be stressed that our M o o n represents the only planetary body of our system, within easy observational reach, to which this argument can be applied. The space missions to Mars, commencing with the historical flight of U.S. Mariner 4 in 1965, have disclosed the surface of the sister planet of ours to be pockmarked with craters of similar kind - and, no doubt, similar origin - as those which have disfigured our satellite. However, unlike for our Moon, the axis of rotation of the Martian globe is considerably more inclined to its orbital plane - the inclination of the Martian equator to the ecliptic now amounts to 23o51 ' (in contrast with 1°32 ' for the Moon), and may likewise oscillate somewhat on account of nutation. Moreover, the oblateness of the Martian globe is bound to give rise to a precession of the Martian axis of rotation, due to solar attraction, with a period of the order of 106 yr which - long as it is in comparison with that of the luni-solar precession of our own planet - is fleetingly short in comparison with the age of the solar system.

COSMIC INFLUENCES ON THE EARLY HISTORY OF THE LUNAR SURFACE

205

Therefore, in the course of its long astronomical past, the directionality of impacts on the Martian surface would have been largely 'de-focussed' by the precession of its rotational axis in space. There is but little r o o m for doubt that cosmic particles, whose collisions with the Martian surface gave rise to such impact craters as we observe there today, were in heliocentric orbits prior to their fatal encounters with our sister planet. As, however, the direction normal to any element of the Martian surface may oscillate by + 2 4 ° in the course of each precessional cycle, the directionality effects of particles in heliocentric orbits could be very largely smoothed out in the course of time. It is the near-perpendicularity of the lunar axis of rotation to the ecliptic which renders our satellite so significant in this c o n n e c t i o n - in addition to many other unique features which it disclosed already to the inquisitive human mind, and which it will no doubt still disclose in the future.

Acknowledgement Prepared at the Lunar Science Institute, Houston, Texas under the joint support of the Universities Space Research Association, Charlottesville, Virginia, and the National Aeronautics and Space Administration under Contract No. N S R 09-051-001. This paper constitutes the Lunar Science Institute Contribution No. 84.

References Clemence, G. M. and Brouwer, D.: 1955, Astron. J. 60, 168. Gault, D. E. : 1972, The Moon, to be published. Koziel, K.: 1967a, Icarus 7, 1. Koziel, K.: 1967b, in Measure o f the Moon (ed. by Z. Kopal and C. L. Goudas), D. Reidel Publ. Co., Dordrecht, pp. 3-11. Marcus, A. H.: 1966, Icarus 5, 165, 178, 190, 590.