Three papers featuring in the current issue of the journal Science describe new findings from the Lunar Reconnaissance Orbiter (LRO) on the Moon's tumultuous history.

Two papers focus on the latest findings from the Diviner Lunar Radiometer Experiment, which reveal previously unseen compositional differences in the lunar highlands. The Moon's geology can very simply be considered as the anorthosite-rich highlands that dominate the farside of the Moon, and the basaltic maria, which infill the massive nearside impact basins. Both units are considered 'primitive', that is, they crystalised from lunar mantle materials. New data from Diviner, however, suggests that the composition of the highlands may not be as clear cut as first thought.

Each rock type absorbs and emits energy in its own unique way, and is measured as a spectral signature at infrared wavelengths by Diviner. Deciphering this signal provides information on how that rock formed. "Diviner is literally viewing the Moon in a whole new light," says Benjamin Greenhagen of NASA’s Jet Propulsion Laboratory, lead author of one of the Diviner papers.

The data clearly shows that some highland regions appear more sodium-rich than that of typical anorthosite crust. This suggests that there may have been variations in the chemistry and cooling rate of magma that formed the crust. The high resolution infrared maps also reveal the presence of highly silicic minerals such as quartz and potassium-rich and sodium-rich feldspar – minerals that are only associated with 'evolved' rocks – those that have undergone extensive magmatic processing. The regions match previously identified areas that exhibit high abundances of the element thorium, another proxy for highly evolved rocks.

"The silicic features we've found on the Moon are fundamentally different from the more typical basaltic mare and anorthositic highlands," says Timothy Glotch of Stony Brook University, lead author of the second Diviner paper. "The fact that we see this composition in multiple geologic settings suggests that there may have been multiple processes producing these rocks."

One mystery that still remains is the apparent lack of evidence for fresh mantle materials, which, rich in iron and magnesium, would be easily detectable by Diviner, and which planetary scientists expect might be exposed in the largest, oldest, and deepest lunar basin, the South Pole Aitken Basin on the farside of the Moon. The fact that no evidence was found by Diviner suggests that no mantle materials are exposed, they may occur in areas too small for Diviner to detect, or it has been mixed up with other materials thanks to later impact events.

Another paper is based on data from LRO's Lunar Orbiter Laser Altimeter (LOLA), from which a detailed global topography map has been created. The findings rekindle a much debated topic in impact cratering that concerns the Moon's bombardment history, and whether there was an apparent spike in the rate of impacts 3.9 billion years ago. Evidence for this spike, a period referred to as the 'late heavy bombardment', was first gleaned from rocks that Apollo astronauts collected, which all seemed to have melted from impacts at the same time. One school of thought suggests this early flurry of impactors came from asteroids pushed out of the Asteroid Belt as the outer gas giants jostled in their orbits, whereas more recent craters were as a result of near-Earth asteroids colliding with the Moon.

Using LOLA, James Head of Brown University and colleagues studied over 5,000 large craters with diameters over 20 kilometres, and found that older regions of the Moon had a greater proportion of large craters than younger regions. Head says this strengthens the argument that there were two different populations of impactors that hit the Moon.

"The highlands have a greater density of large craters compared to smaller ones, implying that the earlier population of impactors had a proportionally greater number of large fragments than the population that characterized later lunar history," he says. "Using the crater counts from the different impact basins and examining the populations making up the superposed craters, we can look back in time to discover when this transition in impactor populations occurred. The LRO LOLA impact crater database shows that the transition occurred about the time of the Orientale impact basin, about 3.8 billion years ago. The implication is that this change in populations occurred around the same time as the large impact basins stopped forming, and raises the question of whether or not these factors might be related. The answers to these questions have implications for the earliest history of all the planets in the inner solar system, including Earth."

But the debate is far from being laid to rest; other scientists say that crater distribution can be misinterpreted because lava flows and debris ejected from other craters – processes which do not occur evenly over the Moon – can cover craters up, giving a biased view of the history of local regions.

A sample return mission to the South Pole Aitken Basin is noted as a high priority target to help solve this debate. As the the Moon's oldest crater, if rocks within it turn out to be older than its 3.9 billion year old age, then the late heavy bombardment theory would be under threat. Studying this large basin in situ will also help constrain theories of the Moon's evolution, and put the Diviner data into context. "The new Diviner data will help in selecting the appropriate landing sites for potential future robotic missions to return samples from SPA," says Greenhagen. "We want to use these samples to date the SPA-forming impact and potentially study the lunar mantle, so it's important to use Diviner data to identify areas with minimal mixing."