The effect of present-day mantle temperature anomalies on crustal thickness inversions for the Moon

Considering a laterally variable crustal thickness has important effects on modeling the 3D geodynamical evolution of terrestrial bodies (e.g., Plesa et al., 2016; Fleury et al., 2024; Santangelo et al., 2025). On the one hand, it provides an orientation for the geodynamic model by correlating subsurface regions with surface features such as craters and volcanic centers. On the other hand, it improves the geodynamic model, allowing it to capture temperature fluctuations induced by thickness variations in a radiogenically enriched and low-conductivity crust.

Asymmetries in the subsurface temperature predicted by geodynamical models at present-day will induce gravity field anomalies that can, in turn, affect crustal thickness inversions. In the case of the Moon, a present-day thermal asymmetry between near- and far-side has been predicted by several studies (e.g., Laneuville et al., 2013, 2018; Park et al., 2025; Santangelo et al., 2025), possibly induced by the concentration of radioactive isotopes underneath the nearside crust. This 100–200 K temperature anomaly in the mantle translates to a large-scale and prominent negative density anomaly, which is yet to be accounted for by inversions of gravity data for the crustal thickness of the Moon (e.g., Wieczorek et al., 2013).

In this work, we couple geodynamic models together with gravity and topography inversions of crustal thickness to provide self-consistent estimates of the lunar mantle and crustal structure. We convert subsurface thermal anomalies predicted by the thermal evolution model into density anomalies using a pressure- and temperature-dependent parameterization of the thermal expansivity (Tosi et al., 2013). The density anomalies are used as input to invert for the crustal thickness distribution. The crustal thickness inversion model used in this study has been adapted from the setup described in Broquet et al., (2024). 

For self-consistency, we iterate between the crustal thickness and the geodynamic model, as the density anomalies obtained in the geodynamic model result from crustal thickness variations and associated distribution of radiogenic isotopes, while the crustal thickness inversion itself depends on the density anomalies and associated density contrast at the crust-mantle boundary. Convergence is reached within a couple of iterations. 

We find that a positive temperature anomaly associated with the enrichment of radiogenic isotopes beneath the lunar near side, as required to explain the Apollo 15 and Apollo 17 heat flux measurements (Langseth et a., 1976), induces a crustal thinning up to 8.5 km in the Procellarum KREEP Terrane (PKT) region. Conversely, the positive density anomaly associated with a colder lunar interior underneath the thin-crust South-Pole Aitken basin produces a crustal thickening of ~3 km.

Our coupled geodynamic crustal thickness models show that the effects of subsurface temperature anomalies can lead to changes in crustal thickness estimates comparable to the uncertainty in the seismically derived crustal thickness measurements (~8 km; Chenet et al., 2006). Thus, considering temperature anomalies on crustal thickness modeling has important implications for our understanding of the crustal structure of the Moon. Upcoming seismic and heat flow measurements will, therefore, be critical to discriminate between different interior structure models.