GLOBAL MAGNESIUM MAPPING ON MARS: INSIGHTS INTO METHODOLOGY-DRIVEN VARIABILITY

Magnesium is a common element in primary igneous and secondary aqueous alteration mineralogy on Mars, especially given a more mafic chemistry compared to Earth’s crust. Consequently, variations in global magnesium concentrations can elucidate martian geologic processes and history, especially as it relates to the extensive archive of interactions between crust and atmosphere within the critical zone. Characterizations of crustal density, groundwater variations at Meridiani, and moderate pH precipitation of magnesite exemplify such potential. However, magnesium is not mapped via Gamma Ray Spectroscopy (GRS) data, and therefore magnesium abundances on a global scale remain largely unconstrained. Furthermore, the relative efficacy and accuracy of computationally deriving Mg, by McSween et al. (2009) and Baratoux et al. (2014), are mostly unknown.

In this study, we compare two key methods to estimate the global variability in Mg mass fraction (wt.%), and stoichiometric MgO wt.%. Our regression-based approach uses Mg/Si vs Ca/Si mass ratios from meteorites to determine a linear relationship. As Ca and Si maps are available, MgO wt.% can then be derived from the regression coefficients. Meanwhile, the mass balance approach yields MgO after all other elements are determined from a combination of chemical maps from GRS and meteoritic ratios for several unreported elements: P, Mn, Na and Ti.

While both considered methods broadly return MgO wt.% values within the expected range for mafic chemistry, divergences yield insight into regional-scale geologic processes. In SW Tharsis for example, we observe a divergence that may favor the presence of Mg sulfates derived from acid-fog, ground ice, and airfall dust alteration as opposed to primary mineralogy. We conclude that while the regression-based approach yields values broadly consistent with primary mineralogy, the mass-balance approach can better characterize MgO variability on Mars. A necessary caveat with the mass-balance approach is that some unusually enriched or depleted regions on Mars violate the model’s primary mineralogic basis; in these cases, a regression-based approach can at least bound the uncertainties. Therefore, a combined application of the two approaches is needed to computationally map Mg on Mars and interpret geologic processes.