The mantle of Mars: Some possible geological implications of its high density

Abstract

The density of the Martian mantle is estimated to be about 3.55 g/cm³ (Reasenberg, 1977). Model mineral assemblages for the Martian mantle (at 30 kbar) were calculated using a modified CIPW norm scheme by adding FeO to model terrestrial mantle compositions. The density of the resulting mineral assemblages vary with increasing FeO content. With pyrolite starting compositions for the terrestrial mantle, the resulting model Martian mantle with density of 3.55 g/cm³ is not garnet-lherzolite like the Earth; rather it is an assemblage properly called oxide-garnet wehrlite: oxide (periclase-wüstite) 2%; garnet 11%; olivine 73%; clinopyroxene 12%; with no orthopyroxene. Partial melting of such an assemblage wouldyield iron-rich, ultrabasic lavas, with extremely low viscosities. Specifically, model partial melts, assuming production from the quaternary eutectic (inferred to be near: op₇ g_{42 cpx43} ox₈) yields an ultrabasic (SiO₂, 41 to 44%) picritic alkali-basaltic melt (norm composition ne 2.5, plag 32, or 2.4, di 20, ol 37, mt 4.4 and ilm, tr), with a computed viscosity of about 12 P at 1200°C. This model for the composition of the Martian surface lavas (derived from geophysical data and petrologic arguments) is in remarkable agreement with a recently published model by Maderazzo and Huguenin (1977) (derived from reflection spectroscopy, experimental and theoretical models for weathering in the Martian environment). The result also appears to be consistent with recent interpretations (Rasool and Le Sergeant, 1977) of Viking atmospheric chemistry results, namely that the Martian crust is potassium poor. There are a number of geological implications which follow, including (1) superfluid lavas may account for some flood and erosional features observed on Mars; (2) the XRF inorganic chemistry experiment on Vikings 1 and 2 (Baird, 1976) indeed may be measuring compositions approaching primary lavas, contrary to current interpretations which favor a rather mature (weathered) soil; (3) ultrabasic (ferrokimberlitic) ash might be a major constituent of the Martian soil, especially if cosmological models concerning the incorporation of a much volatile material within the early accreting Mars are correct—a matter of current debate; (4) a number of mineral assemblages not previously considered are possible in the Martian mantle depending principally on the activity of volatile substances, (S, O, C, H); it is possible that some very unusual magmas are produced on partial melting; and (5) some ferro-granite melts might be produced by liquid immiscibility.