52nd Lunar and Planetary Science Conference 2021 (LPI Contrib. No. 2548) 1132.pdf

THE ROLE OF ASSIMILATION AND FRACTIONAL CRYSTALLIZATION IN EVOLVED MARTIAN CRUSTAL COMPOSTIONS. A. M. Ostwald1, A. Udry1, E. Gazel2, V. Payré3, P. Wu2, 1University of Nevada, Las Vegas, 4505 S Maryland Pkwy, Las Vegas NV 89154; [email protected], 2Cornell University, 2122 Snee Hall, Ithaca, NY 14853, 3Northern Arizona University, 624 Knoles Dr., Flagstaff, AZ 86011.

Introduction: The martian crust, once thought to be composed of tholeiitic basalt, has proven to be compositionally diverse [1,2]. Recent findings of felsic (>55 wt.% SiO2) targets at Gale crater likely result from fractional crystallization [2-5]. Although fractional crystallization alone can account for these felsic compositions [2], some felsic targets might only form under high degrees of fractionation (>60%), when residual melt is difficult to remove. One process common to Earth that has yet to be investigated in this context for Mars is assimilation and fractional crystallization (AFC). Here we discuss the role of assimilation in the formation of felsic rocks on Mars, and of the diversification of the Mars crust. Methods: We used the Magma Chamber Simulator (MCS) to conduct AFC models. The MCS models the Fig. 1: Model parameters used in this project listed by relevant interactions between a chemically evolving magma, a subsystem. Schematic depicts MCS model of AFC.

cumulate pile, and a wallrock (assimilating crust) [6]. In Model Results: Assimilation in 1 kbar pressure MCS AFC models, magma intrudes the colder wallrock. systems occasionally begins late during crystallization The exothermic process of fractional crystallization (~60% solid cumulates), and lasts for shorter overall heats the wallrock, which may partially melt, change duration. The duration of assimilation lasts longer and composition through melting, and mix into the magma begins earlier for deep-seated, high pressure systems body (Fig. 1) [6]. All MCS models rely on a MELTS- (4–6 kbar). Melts in higher pressure models (4–6 kbar) family algorithm, and here we selected Rhyolite- are more alkaline-enriched than melts in lower pressure MELTS v.1.0.2 to approximate a hydrous, silica- models. In all models, increasing the water content (to enriched magma in low-pressure environments [7]. 1.0 wt.% H2O) decreases the melt alkalinity relative to Model parameters and starting compositions. We water-depleted (0.07 wt.% H2O) models, but this effect calculated isobaric models at varying pressures (1, 2, 4, is less pronounced as pressure increases. and 6 kbar) to represent depths of the shallow The melting temperature of NWA 7034 was ~50° C subsurface to the base of the average thickness of the lower than that of the average Mars crust, causing it to Mars crust [2,8]. The oxygen fugacity (fO2) in each melt earlier and for longer duration in models where it model was constrained at the Fayalite-Magnetite-Quartz is the wallrock subsystem. Fastball starting composition (FMQ) buffer. We used the near-primary compositions has a slightly more alkaline starting composition than of Fastball and an average of the Adirondack basalts the averaged Adirondack-class basalts, and the former (both observed in Gusev crater) as the initial magmatic evolves to be more alkaline throughout the models. composition, to which we added initial water contents Assimilation of NWA 7034 wallrock resulted in slightly of 0.07, 0.5, and 1.0 wt.% [9,10]. For wallrock, we used more alkaline compositions overall than did the bulk rock composition of regolith breccia meteorite assimilation of the average Mars crust. Northwest Africa (NWA) 7034 and an estimate for the Model Fits: Best model fits are presented against average Mars crust [11,12]. Wallrock starting evolved Gale crater targets, the compositions of which temperatures were calculated along a 15°C/km are found in [14,15]. Overall, felsic targets were better areotherm (martian geothermal gradient) [13]. These reproduced by lower pressure (1–2 kbar) and more parameters and all additional parameters (not presented hydrous (0.5–1.0 wt.% H2O) magmas. Alkaline targets here) are detailed in Fig. 1. were better reproduced by higher pressure (4–6 kbar)

systems with less hydrous (0.07 wt.% H2O) compositions (Fig. 2). 52nd Lunar and Planetary Science Conference 2021 (LPI Contrib. No. 2548) 1132.pdf

Fastball starting composition. Fastball assimilating NWA 7034 wallrock at fits alkaline target Harrison d (basaltic trachyandesite) at 54% crystallinity (Fig. 2a). Similarly, Fastball assimilating the average Mars crust at 2 kbar and with 0.50 wt.% H2O fits Harrison within 57% crystallinity (Fig. 2b). Adirondack basalt starting composition. The Adirondack class basalt assimilating NWA 7034 wallrock at 1 kbar and 1.0 wt.% H2O fits an andesitic Gale crater target (Sparkle) at 57% crystallinity. At 4 kbar and 1.0 wt.% H2O, the model fits Harrison at 55% crystallinity (Fig. 2c).

a Fig. 2: (a) Fastball magma assimilating NWA 7034. (b) Fastball magma assimilating average Mars crust. (c) Adirondack-class basalt assimilating NWA 7034. (d) Adirondack-class basalt assimilating average Mars crust. Dashed line is from [17]. Discussion: Higher pressures and higher initial temperatures of wallrock result in more assimilation, as the wallrock closer to its solidus melts more readily and for longer throughout the model. In the case of lower pressure models (1–2 kbar), where there is less assimilation overall, target evolved compositions are achieved within low degrees of crystallinity because of the addition of liquid at later stages of magmatic evolution via wallrock melting. Higher pressures typically result in increased alkalinity in fractional b crystallization, but highly alkaline compositions are feasible at 4–6 kbar with significant additions of incompatible elements from the melting wallrock. The controlling factor overall on assimilation using Mars-relevant compositions and magmatic conditions is the initial temperature of the wallrock, which is dictated by the aerothermal gradient temperatures. Thus, assimilation was likely a more common process on early Mars, which was hotter than it is today [16]. Assimilation would have become a less prevalent process, or was restricted to high pressures, as the planet cooled. These observations may help to resolve the temporal contrasts in martian crustal diversity, whereby c older terrain tends to contain more diverse (and evolved) igneous compositions than does younger terrain and young martian meteorites. References: [1] McSween H. Y. et al. (2009) Science, 325, 736- 739. [2] Udry A. et al. (2018) JGR, 123, 1525-1540. [3] Stolper E. M. et al. (2013) Science, 341, 6153. [4] Udry A. et al. (2014) JGR, 119, 1-18. [5] Sautter V. et al. (2015) Nat. Geosci., 8, 605-609. [6] Bohrson W. A. (2014) Journ. Pet., 55, 1685-1717. [7] Gualda G. A. R. et al. (2012) Journ. Pet., 53, 875-890. [8] Wieczorek M. A. and Zuber M. T. (2004) J. Geophys. Res., 109, E01009. [9] Filiberto J. R. et al. (2010) Geophys. Res. Lett., 37, L13201. [10] Squyres, S. W. et al. (2007) Science, 316, 738-742. [11] Agee C. B. et al. (2013) Science, 339, 780-785. [12 ] Taylor S. R. and McLennan S. M. (2009) Cambridge Uni. Press. [13] Hahn B. C. et al. (2011) LPSC XLII, Abstract #2340. [14] Cousin A. et al. (2017) Icarus, 288, 265-283. [15] Schmidt M. E. et al. (2014) JGR, 119, 64-81. [16] Elhmann B. L. et al. (2011) Nature, 479, 53-60. [17] Irvine T. N. J. and Baragar, W. R. A. F. (1971) Canadian Jour. Earth Sci., 8, 523-548.