A reassessment of the emplacement and erosional potential of turbulent, low-viscosity lavas on the Moon

We have reevaluated the role of thermal erosion by low-viscosity lunar lavas as a mechanism for the formation of the lunar sinuous rilles. We have adapted the model of Williams et al. [1998] and used the compositions of an Apollo 12 basalt and a terrestrial komatiitic basalt to investigate the compositional and environmental effects on the flow behavior of low-viscosity lavas on the Moon and the Earth. Our model predicts that lunar lavas could have erupted as turbulent flows that were capable of flowing hundreds of kilometers on a sufficiently flat, unobstructed substrate. These results are consistent with previous studies. Modeling of lavas over a substrate of the same composition shows that thermal erosion rates would have been low (∼10 cm d^-1). As a result, long-duration eruptions (approximately months to years) would have been required to incise deep (tens to hundreds of meters) channels. Partial melting and mechanical removal of the substrate, a mechanism suggested by Hulme [1973] to enhance erosion, only slightly increases thermal erosion rates. Other factors, such as higher flow rates or lava superheating, could have produced deep rilles by thermal erosion during shorter-duration eruptions. A superheated lunar lava not only would have had a higher erosion rate (∼40 cm d^-1) but also would have remained uncrusted for tens of kilometers, which is consistent with the open channel morphology of most sinuous rilles. For lunar lavas with large volatile (i.e., vesicle) contents, the presence of vesicles would have tended to increase viscosity at low strain rates, resulting in shorter turbulent flow distances, lower thermal erosion rates, and thus shallower erosion channel depths for given eruption durations.