The rugged highlands of the lunar farside may be the remains of a smaller companion moon that sunk into its surface, according to new computer simulations conducted by planetary scientists at the University of California, Santa Cruz.

The dichotomy between the near and farsides of the Moon has long been debated: why is the nearside relatively low-lying with giant basins infilled with lavas, whereas the farside is dominated with mountainous terrain and a crust up to several tens of kilometres thicker?

It is a well-accepted theory that the Moon was formed from the debris of a giant impact of a Mars-sized object with the Earth early in its history. This cataclysmic impact could have also spawned other shorter-lived moonlets, and in the new research one such moon – approximately one-thirtieth the mass of the Moon – spirals into the lunar surface at the relatively low velocity of 2.4 kilometres per second. Instead of blasting a crater and excavating material that is typical of much higher velocity impact events, the collision added material to the Moon.

“Of course, impact modelers try to explain everything with collisions,” says one of the paper's authors, Erik Asphaug. “In this case, it requires an odd collision: being slow, it does not form a crater, but splats material onto one side. It is something new to think about.”

The candidate moonlet was likely stable for tens of millions of years after the Giant Impact in a gravitationally stable "Trojan" point in the Moon's orbit, only to be destabilized after the Moon's orbit had migrated away from Earth.

“One of the nice aspects of our theory is that it works very well with the Moon-forming giant impact theory, which naturally can lead to more than one Moon forming in the massive debris disc,” Martin Jutzi tells Astronomy Now.“In our study we propose that there was only one such collision leading to the farside highlands.”

The new model can also explain compositional variations in the lunar crust. Nearside crust is rich in potassium, rare-earth elements, and phosphorus (collectively known as KREEP), and uranium and thorium. These elements are thought to have been the last of a once global magma ocean to crystallize into rock. In the simulations, the collision forces the KREEP-rich layer onto the opposite, nearside hemisphere, creating the dichotomy seen today.

There are many other plausible explanations for the contrasting hemispheres, though, ranging from the piling up of material from the impact that created the 2,500 kilometre-wide South Pole Aitken Basin on the far side of the Moon, to variations in tidal forces and internal heating and dynamics, and impact bombardment and shallow melting.

“Sample return missions would certainly help to test our theory,” says Jutzi. “In our scenario, the impacting companion moon has a slightly older crust. Therefore it would be very interesting to have samples from the Moon's farside.”

The scientists explain in their paper, published in the 3 August issue of Nature, that although the companion moon would have condensed from the same material as the Moon, and thus share compositional similarities, its lesser size meant it coalesced faster, and was therefore older than the Moon, a signature that may still be preserved.

Ian Garrick-Bethell, also of UCSC but a proponent of the tidal heating theory, likes the new idea but thinks “more work should be done to understand if the observed topography, crustal thickness, and composition of the far side highlands are truly consistent with the model predictions.”

There might not be too long to wait. “The GRAIL mission will provide detailed information about the crustal thickness,” says Jutzi. “A comparison of our model to this data set will be a test of our theory.”