The farside of the Moon has always been a mystery and is only accessible by spacecraft. New compositional information from the Moon Mineralogy Mapper (M3) onboard Chandrayaan‐1 has identified a suite of highly unusual rock types exposed at small areas within the farside Moscoviense Basin. M3 is a state‐of‐the art visible and near‐infrared imaging spectrometer that was a guest instrument on Chandrayaan‐1, the Indian Space Research Organization’s (ISRO) first mission to the Moon. The instrument is designed to measure accurately the diagnostic mineral absorption bands of solar radiation reflected from the lunar surface.
The instrument performed exceptionally well. The M3 data presented and discussed here are extracted from an orbit of data taken early in the Chandrayaan‐1 mission on 25 January 2009, with nearly a full orbit of M3 data acquired across the western rim of the Moscoviense Basin.
Five unusual regions exhibited remarkably consistent spectra, all quite different from local lithologies. They can be divided into three distinct rock type groups dominated by the following mafic minerals: (1) orthopyroxene, (2) olivine, and (3) Mg‐Al spinel. This family of unusual rock types was designated as OOS. All are highly enriched in the dominant mafic mineral present.
Identified along the innermost Moscoviense Basin ring are several small exposures of three separate but distinctive rock types, the OOS, which contain an exceptionally high concentration of orthopyroxene, olivine, and Mg‐rich spinel, respectively. The OOS rocks and developed soil have remained in place undisturbed along the Moscoviense ring since the basin formed.
For the Moscoviense Basin, we believe the OOS represent components of the lower crust. The thin to nonexistent crustal thickness estimates for the region suggest that the OOS zone of origin may even approach the crust‐mantle interface. The three OOS lithologies are very distinctive and each occurs in more than one location. At the spatial scale of M3 no clear gradients or mixing between OOS are observed. Although the OOS are widely dispersed along the inner ring, we have no direct information about any relationship between the three lithologies. All fade into background material within several pixels. There are cases where each appears completely separately from the others. On the other hand, there are cases where the spinel lithology is spatially close to the orthopyroxene lithology and where the orthopyroxene lithology occurs in close proximity to the olivine lithology, suggesting they may be genetically linked.
We propose that the OOS are differentiation products of one or more plutonic events that intruded magmatic material into the lower part of the extensive feldspathic crust, itself derived from the magma ocean. The small size of the OOS (a few kilometers) suggest they could represent several separate small plutonic events or a large cooled plutonic body disrupted by the basin forming event.
We are confident of the characterization of the mineral compositions of the OOS. The position of OOS on the innermost ring of the Moscoviense Basin is evidence of their sampling from great depth. Their mineralogy is not consistent with upper crustal anorthositic material sampled by other basins, but rather strongly suggests that the Moscoviense Basin sampled down to lower crustal material. With the exception of the Mg‐spinel lithology (which is new), the orthopyroxene‐ and olivine‐rich mineralogy seen in OOS rock types are not unknown on the Moon. But they have never been seen in their actual geologic setting, nor in the concentrations implied by the M3 data for OOS areas. Furthermore, as the remaining M3 data are calibrated, additional outcrops of the Mg‐spinel lithology have been also found at one other basin, confirming that this new rock type plays a significant role in lunar crustal structure.
This information provides important new insights and constraints on the character and evolution of early planetary crusts. This new compositional information about the lunar crust also opens new avenues of inquiry that are beyond the scope of the discussion presented here. What is the initial composition of melts that are needed to produce the three lithologies? How and when was the melt formed? What materials were melted? Where? How did the mineral separation and concentration occur at the scale observed? What size magma chamber is needed to allow such clear separation of lithologies as a layered intrusive? What depth? What temperature?
These OOS results also provide a taste of major surprises that come from probing and analyzing data acquired by modern sensors orbiting the Moon. We have barely scratched the surface.
(a) Olivine‐anorthite‐silica pseudoternary commonly used to describe crystallization sequence for highland rocks. A trend of fractional crystallization involving olivine, anorthite, and pyroxene for several example initial melt compositions (X1, X2, X3, X4) is depicted. Since spinel is not an end‐member of this pseudoternary, an arrow indicates the approximate compositional direction of spinel crystallization trends. The origins of possible melt compositions within the ternary have yet to be constrained (Mg‐ and Fe‐ rich are on the left, Al‐rich on the right). OL, ol, olivine (Mg, Fe)2SiO4; An, an, anorthite CaAl2Si2O8; Si, si, silica (SiO2); Opx, opx, orthopyroxene (Mg, Fe)2SiO3; sp, spinel (Mg, Fe)Al2O4; PSA, pink spinel anorthosite; TROC, troctolite. (b) Schematic cross section for the Moscoviense region illustrating possible components of the deep lunar crust prior to the basin impact. The upper crust is dominantly anorthosite overlain by a megaregolith. The lower crust is more mafic. It may contain diverse magmatic plutons that exhibit a variety of compositions, several of which are observed as outcrops at Moscoviense. Although the relative portions of minerals present are unknown, several OOS exposures are highly enriched with mafic minerals, suggesting strong concentration. Possible starting compositions that might produce the mineralogy in these hypothetical plutons include: A (X2), B (X3), C (X4), D (X1 with crystallization before the melt reached the olivine field), E and F (X1 with two sequences of melt‐crystal separation).