Long-lived lunar volcanism sustained by precession-driven core-mantle friction

Abstract

Core-mantle friction induced by the precession of the Moon's spin axis is a strong heat source in the deep lunar mantle during the early phase of a satellite's evolution, but its influence on the long-term thermal evolution still remains poorly explored. Using a one-dimensional thermal evolution model, we show that core-mantle friction can sustain global-scale partial melting in the upper lunar mantle until ∼3.1 Ga, thus accounting for the intense volcanic activity on the Moon before ∼3.0 Ga. Besides, core-mantle friction tends to suppress the secular cooling of the lunar core and is unlikely to be an energy source for the long-lived lunar core dynamo. Our model also favours the transition of the Cassini state before the end of the lunar magma ocean phase (∼4.2 Ga), which implies a decreasing lunar obliquity over time after the solidification of the lunar magma ocean. Such a trend of lunar obliquity evolution may allow volcanically released water to be buried in the lunar regolith of the polar regions. As a consequence, local water ice could be more abundant than previously thought when considering only its accumulation caused by solar wind and comet spreading.

Keywords: Moon; orbit; precession; volcanism.

Figure 1.

Time evolution of the semi-major axis for (a) the nominal lunar orbit according to [15], and for parameterised orbits with (b) a_s = 34–40 R_E and Γ = 2.0 and (c) a_s = 36 R_E and Γ = 0.5–3.0, where R_E is the Earth’s radius. The dashed black line marks the semi-major axis for the transition to the current Cassini state. The grey box specifies the lunar magma ocean (LMO) phase ending at ∼4.2 Ga.

Figure 2.

(a) Comparison between the present-day temperature at the bottom of the stagnant lid (plus symbols) and upper mantle temperature (circles) of all successful cases. The gray area indicates the range of possible temperature profiles of the lunar mantle over a depth of 0–1200 km inferred from Apollo seismic data [27]. Purple curves specify the range of the lunar mantle temperature inferred from the electrical conductivity data by assuming a dry mantle with its typical composition [29]. Pink and gold curves denote mantle temperatures inferred from electrical conductivity data by assuming a wet mantle with a bulk water content of 100 ppm and a dry mantle, respectively [28], with solid and dashed lines referring to an olivine (solid) and orthopyroxene (dashed) composition, respectively [28]. (b) Time evolution of the temperature at the crust-mantle boundary for the successful cases. The black bar indicates the critical temperature of 1300 K corresponding to the transition in relaxation of pre-Nectarian basins [30]. (c) Time evolution of the mantle temperature and (d) of the frictional power for the successful cases.

Figure 3.

Time evolution of the stagnant lid thickness (dash–dot lines) and partial melt region (contour) for (a) a wet-Moon case with η_r = 10²¹ Pa s, a_s = 37 R_E, Γ = 2.0 and ΔT_m = 90 K, and (b) a dry-Moon case with η_r = 10²¹ Pa s, a_s = 38 R_E, Γ = 1.0 and ΔT_m = 180 K. The black dashed line indicates the thickness of the primordial crust, which is held fixed throughout the simulations. The solid and dashed red lines indicate the depths of the upper thermal boundary layer and the bottom of the partial melt region, respectively. The coloured contour represents the volumetric melt fraction in the partially molten mantle. In panel (a), the maximum depth of the melt region is 542.70 km and is attained at 3.94 Ga, with mantle melting ending at 3.06 Ga. In panel (b) the maximum depth of the melt region is 541.80 km and is attained at 3.70 Ga, with mantle melting ending at 3.08 Ga

Figure 4.

(a) Time evolution of the CMB temperature for all successful cases. The dashed line notes the solidus temperature of the lunar mantle at the CMB. The cut-off of the CMB temperature relates to the percolation of the melt when the core-mantle friction is vigorous enough to remelt the bottom of the lunar mantle. (b) Time evolution of the heat flux q_cc for all successful cases. The positive q_cc indicates the activation of a core dynamo.