image: The lunar magma ocean (LMO) cooled rapidly to form early mantle cumulates before tidal locking, and floating plagioclase formed the anorthositic crust. After tidal locking was established, the nearside-farside thermal contrast triggered lateral convection within the LMO, which moved anorthositic crust toward the farside and residual melts to the nearside. Continuous cooling of residual melts ultimately produced thinner late-stage LMO cumulate layers on the farside. N, nearside; F, farside.
Credit: Yuantao Gu, Hejiu Hui
The research team, led by Professor Hejiu Hui from Nanjing University, has made a breakthrough in deciphering the origin of the Moon's nearside-farside dichotomy. Using samples returned from the lunar farside by the Chang'e-6 (CE6) mission, the team performed detailed geochemical analyses of anorthosites, Mg-suite rocks, and mare basalts. These farside samples were systematically compared with nearside samples from the Apollo and Chang'e-5 missions. Their study concludes that lunar dichotomy originated from lateral thermal convection within the lunar magma ocean (LMO) after the Moon became tidally locked to Earth.
The Moon’s nearside and farside exhibit striking geomorphological and compositional contrasts. The nearside is dominated by extensive mare basalts and elevated abundances of potassium, rare earth elements, and phosphorus (KREEP), whereas the farside consists predominantly of ancient highlands terrane, characterized by a thicker crust and only sparse mare basalts. Over the past decades, various hypotheses have been proposed to explain this dichotomy, including asymmetric magma ocean convection, giant impact, tidal dissipation, and asymmetric crustal growth. These models remain contentious, and the origin of the lunar dichotomy is still vigorously debated. Moreover, observations of nearside-farside asymmetry have been largely limited to the surface and crust; whether the asymmetry propagates into the lunar mantle has long remained an open question.
In 2024, the CE6 mission accomplished the first sample return from the lunar farside, enabling direct sample-based inter‑hemispheric comparison. CE6 basalts preserve depleted mantle source signatures comparable to those of Apollo low‑Ti basalts, indicating no substantial compositional difference in mantle cumulates between the nearside and farside. In addition, Mg‑suite rocks form via crust‑mantle interactions driven by partial melting of mantle cumulates, so their magnesium number (Mg#) provides a proxy for mantle composition. The comparable Mg# between CE6 and Apollo Mg-suite samples confirms that mantle cumulates are compositionally similar across both hemispheres. Collectively, these findings demonstrate that mantle cumulates crystallized from the LMO exhibit no significant nearside-farside compositional asymmetry.
In contrast, the lunar crust exhibits pronounced hemispheric compositional differences. CE6 anorthosites yield higher peak Mg# values than Apollo anorthosites, pointing to a more magnesian farside crust — consistent with remote‑sensing observations. Meanwhile, CE6 Mg‑suite rocks contain very low rare‑earth‑element concentrations and lower Th/Sm ratios than their Apollo equivalents. This pattern implies either that farside Mg‑suite magmatism operated independently of KREEP, or that the KREEP‑rich layer was distributed asymmetrically between the two lunar hemispheres.
Building on these mantle‑to‑crust systematic comparisons, the team proposed a new model for the origin of lunar dichotomy. Prior to tidal locking of the Moon, the early magma ocean cooled and solidified symmetrically, producing the early mantle cumulates and anorthositic crust. After tidal locking was established, when Earth and the Moon were much closer, tidal heating on the nearside exceeded that on the farside. The resulting inter‑hemispheric temperature gradient drove large‑scale lateral thermal convection: hot magma flowed from the nearside to the farside, transporting early-formed, magnesium-enriched anorthosites to the farside. This mass transport produced a thicker, more magnesian crust on the farside. Concurrently, residual melts were advected toward the nearside, forming only thin KREEP and ilmenite-bearing cumulates (IBC) layers on the farside. This lateral thermal convection model offers a self-consistent explanation for the hemispheric variations observed from mantle to crust, advancing our understanding of the Moon's early evolutionary history.