Timescales of solidification, crystal sedimentation and cumulate compaction
Authors: Ch.-Ed. Boukaré, Marc Parmentier and Steve Parman
At the end of planetary accretion, magma ocean (MO) evolution is thought to set the initial conditions for the long-term evolution of terrestrial planets. Most aspects of MO dynamics are derived from the lunar MO based on data of the Apollo mission obtained more than forty years ago. However, crucial aspects of MO evolution are still highly debated.
One of the key aspects of MO ocean evolution are the time of MO solidification, the degree of silicate differentiation and the initial degree of mantle mixing. If interpreted as the first solid lunar surface, the relative young age of the lunar anorthosite must be reconciled with a relative fast MO solidification suggested by the canonical thermal models. The fractional solidification and the overturn hypothesis, while only an ideal limiting case, can explain important geochemical features of both the Moon and Mars. However, the actual geodynamic evolution of the lunar mantle from a partially (or fully) MO to a solid mantle remains poorly understood. Consistent geodynamic scenarios are nevertheless important as they would provide consistency between the timescales of the physical processes involved in MO ocean evolution such as crystal sedimentation, cumulate compaction, MO time of solidification and overturn timing.
We propose a 1D-sherical model that links in a self-consistent fashion these various timescales. The model couples a thermal budget, a mechanical description of crystal settling in vigorously convective fluid and a compaction model for modeling the melt extraction from the cumulate.
At this stage, this work shows two interesting outcomes. First, imposed MO surface heat flux does not dictate the timescale of MO solidification if convective heat transport in the MO cannot sustain this surface heat flux. This model indicates that this scenario can occur if the MO convective vigor is decreased due to an increase of the suspended crystal fraction. The suspended crystal fraction in the MO increases if the timescale solidification is shorter than the timescale of crystal sedimentation. Second, by linking crystal settling and cumulate compaction, this approach shows that only a very small suspended crystal fraction allows time for cumulate compaction during MO solidification. Indeed, a low suspended crystal fraction appears to be the most likely mechanism that can decrease sufficiently the sedimentation rate that cumulate compaction is faster than cumulate accumulation. As shown in previous work, solid-state-like convection in the cumulate can start during MO solidification if the effective viscosity of the cumulate is low enough or if the time of MO solidification is long enough. By constraining the amount of retained melt fraction in the cumulate with the compaction model, this work allows to better estimate the cumulate viscosity and thus provides new constrains for the issue of very early convective mixing in terrestrial mantle.
One of the key aspects of MO ocean evolution are the time of MO solidification, the degree of silicate differentiation and the initial degree of mantle mixing. If interpreted as the first solid lunar surface, the relative young age of the lunar anorthosite must be reconciled with a relative fast MO solidification suggested by the canonical thermal models. The fractional solidification and the overturn hypothesis, while only an ideal limiting case, can explain important geochemical features of both the Moon and Mars. However, the actual geodynamic evolution of the lunar mantle from a partially (or fully) MO to a solid mantle remains poorly understood. Consistent geodynamic scenarios are nevertheless important as they would provide consistency between the timescales of the physical processes involved in MO ocean evolution such as crystal sedimentation, cumulate compaction, MO time of solidification and overturn timing.
 |
| Caption: Sketch of segregation processes occurring in magma oceans (MOs). This study investigates the links between the timescales of MO solidification and crystal sedimentation. It is here proposed that high suspended crystal affects the MO solidification timescale. |
We propose a 1D-sherical model that links in a self-consistent fashion these various timescales. The model couples a thermal budget, a mechanical description of crystal settling in vigorously convective fluid and a compaction model for modeling the melt extraction from the cumulate.
At this stage, this work shows two interesting outcomes. First, imposed MO surface heat flux does not dictate the timescale of MO solidification if convective heat transport in the MO cannot sustain this surface heat flux. This model indicates that this scenario can occur if the MO convective vigor is decreased due to an increase of the suspended crystal fraction. The suspended crystal fraction in the MO increases if the timescale solidification is shorter than the timescale of crystal sedimentation. Second, by linking crystal settling and cumulate compaction, this approach shows that only a very small suspended crystal fraction allows time for cumulate compaction during MO solidification. Indeed, a low suspended crystal fraction appears to be the most likely mechanism that can decrease sufficiently the sedimentation rate that cumulate compaction is faster than cumulate accumulation. As shown in previous work, solid-state-like convection in the cumulate can start during MO solidification if the effective viscosity of the cumulate is low enough or if the time of MO solidification is long enough. By constraining the amount of retained melt fraction in the cumulate with the compaction model, this work allows to better estimate the cumulate viscosity and thus provides new constrains for the issue of very early convective mixing in terrestrial mantle.
Related papers:
EGU Vienna, 2019
LPSC Houston, 49 and 50
AGU Fall meeting, 2018 (Invited Talk)