Joe McCall has doubts about the Giant Impact Theory (GIT) about the origin of the Earth-Moon system.
Geoscientist Online 20.05 April 2010
I have previously expressed intuitive doubts about the Giant Impact ‘theory’ for the origin of the Earth/Moon system1. This ‘theory’ required, at the time I was writing, the impact on the proto-Earth of a body three times the size of Mars, in a grossly errant orbit. Subsequently, in 2009, Melosh2 virtually demolished this ‘theory’, which was based on computer modelling, and satisfies the rotational characteristics of the Earth/Moon system, and therefore astrophysicists. In its latest form, it predicts that 70% of the Moon’s mass was contributed by the projectile, whereas only about 10% of the Earth’s mass was so contributed2. Melosh’s argument is that, over the past decade, an increase in precision of isotopic measurements, particularly for oxygen, has shown that the Earth and Moon are indistinguishable form one another at a level of 5 parts per million. Unless the isotopic compositions of proto-Earth and projectile were nearly identical by some fortuitous coincidence, there should be detectable differences, as argued by Pahlevan and Stevenson3. These authors suggested that turbulent convection could have equilibrated these values, but Melosh discounted this escape from the problem2. Melosh concluded that this has brought us to a new crisis in the still unresolved problem of the Moon’s origin. New bold ideas will be required in the future.
I have recently published4 a challenging article suggesting that Plate Tectonics, as we know it, did not operate in the Archaean and that, early on, there was dispersed eruptivity piercing the sialic crust of the Earth from plumes, forming short rifts, but there was no longitudinal continuity to these as in the case of the present cycle mid-ocean ridge spreading centres, and so large continental plates could not separate and come into existence. My essential point was that it is mistaken to apply the dogma of the present spreading plate tectonic cycle (mid-Jurassic to present) to the Archaean. I envisaged a possible stepwise change in the convection patterns in the Mantle from cycle to cycle, one such step occurring at the end of the Archaean. Though it is not my main contention, I suggested that the Archaean Earth was essentially a uniplate planet, with a globe-enveloping sialic crust. My most ‘way-out’ suggestion was that - contrary to the conclusion of Taylor and McLennan5, that despite the recognition of evidence of sialic crust way back to 4360 Ma from the Jack Hills, Western Australia6, that none then existed - the globe encircling sialic envelope stemmed from an original magma ocean formed before this date, such as is accepted for the origin of the primordial anorthositic covering of the Moon5. The difference in composition was attributed by me to the larger Earth retaining its volatiles, whereas the Moon did not.
What I tentatively suggested about the globe encircling sialic envelope being initiated at the Earth’s very beginning as the top of a magma ocean, such as is invoked for the early Earth as well as the Moon by Taylor and McLennan5, seems consistent with a hypothesis that the Earth and Moon formed as separate bodies, likely as binary system, without the intervention of a Giant Planet collision. The concept of about a dozen wandering planets (three of which impacted Mercury, the Earth and Mars) appears three times in Taylor & McLennan’s book5, yet appears to be little more than supposition, derived from computer modelling and found to be ‘compatible with the rotational characteristics of the Earth/Moon system’1. ‘A model is not evidence’ (Plimer7). There appears to be no real evidence at all that these wandering planets ever existed at the time of the proto-Earth, yet they are nowadays treated almost as fact?
My intuitive doubts arose that the Giant Impact Theory requires a collision between the Earth and a planet on an errant orbit virtually at the very beginning of planetary history in the solar system timewise - we have dates as far back as 4360 million years from the Jack Hills zircons on Earth6 and the anorthositic rocks of the Moon have yielded dates about as far back, so we know how long ago Earth and Moon were separate. The separation must have been well before these age dates. The planets accreted from planetesmals derived from the Solar nebula, and the nebula from which they accreted seems to have already been oriented in a plane around the sun, so as to station the accreting planets in the plane of the ecliptic.
Michael Mayor, the leading authority on exoplanets, (pers. comm.) concludes that “we have serious reasons to believe that in multiplanetary systems, the planets are almost coplanar. However, the discovery of planets with elongated orbits points to the possibility of important gravitational scattering between massive planets, a phenomenon that will not preserve coplanarity. It has recently been discovered that a large fraction of short-period exoplanets have orbits at quite a large angle with the stellar equatorial plane, and also planets with retrograde orbits due to the so-called Kozai effect (dynamical interaction between planets), leaving us with the conclusion that a lot of multiplanetary systems are coplanar, but it is not a rigid rule”.
This evidence might seem to favour the Giant Impact Theory. However, we are concerned with a single solar system, in the case of the Giant Impact Theory. Taylor and McLennan invoke no less than three errant planet collisions with planets on the ecliptic, Mercury, Earth and Mars. Only a minority of the total of errant planets, if the existed at all, could by chance have been on such collision courses: logical thinking demands that there must have been a huge number of planets on errant orbits in the early solar system, for three such collisions to have occurred at all, and the majority of these should still be in existence. Where are these errant planets? I believe that besides Melosh’s objections based on isotopic evidence, rational thinking is against the Giant Impact Theory.
Can we not go further than Melosh’s admirable final statement, and, rather than try to find new collision models of giant impacts by other techniques as he suggests, consider whether the Moon is not after all a proper satellite like the Galilean satellites of Jupiter? Pluto, too, like the Earth, also has an overlarge single satellite in Charon, the oversize of the Moon satellite is not unique in the solar system.
I can see a possible analogy with the dumbbell tektite (Figure 1), hurled molten into space from huge terrestrial impact8, and which was liable to separate into two droplet tektites. Could not the proto-Earth/Moon have separated in the same way from a spinning partly-molten proto-body, with unequal ends to the dumbbell? The two separated piece that originally formed the spinning dumbbell would have adjusted to spherical shape, just as Ceres, the largest asteroid has. This occurs in all bodies a certain size threshold.
Scientists of the future will predictably, if this hypothesis is scientifically acceptable, need to invent a quite new explanation to get around the known objections related the Earth/Moon motions to such a solution of the Earth/Moon problem.
Space scientists seem to turn to impact to explain everything that they cannot understand: impact has been an important process in moulding the Earth’s surface in the past9, but seems to be a bit of an obsession to the present generation of planetologists? It seems likely that future generations will see an overemphasis on impact processes in the Extraterrestrial Science of our age. Likewise, as Carlyle10 in the early nineteenth century regretted that the age of technology and measurement was taking over from the inductive approach of the age of William Herschel, this age may be deplored as age of computer modelling dominating science: such modelling does not replace the older methods of science, and can be overturned.