Sorry, you need to enable JavaScript to visit this website.


Tabb Prissel
Abstract Text: 
Troctolites: Two competing hypotheses suggest melts parental to ancient lunar troctolites formed (I) by partial melting of a hybridized mantle source region comprised of ultramafic cumulates + plagioclase-bearing rocks + KREEP [1] or (II) when primitive mantle-derived MgO-rich (and plagioclase-undersaturated) melts became contaminated with the lunar crust (+/- KREEP) during magma-wallrock interactions [2]. To further constrain the models, a combination of phase equilibria and melt-rock reaction experiments have been performed on a range of Mg-suite parental melt compositions to investigate the production of two distinct spinel populations within lunar troctolites [3] — chromite-bearing (FeCr2O4) troctolites and dunites, and the rarer pink spinel (MgAl2O4) troctolites (PST). Phase equilibria experiments testing high-Al plagioclase-saturated magmas from model (I) exclusively yield major and accessory mineral compositions consistent with the rarer PST, but do not yield chromite [3]. Recent melting experiments testing plagioclase-bearing Mg-suite source regions also yield PST mineralogy [4], further supporting the phase equilibria experiments from [3]. Instead, melt-rock reaction experiments produce a range of accessory spinel compositions: pink spinel forms near the high-Al contaminated melt-rock interface, whereas chromite stabilizes in the low-Al less-contaminated regions away from the melt-rock interface [5]. In order to explain both chromite- and pink spinel-bearing troctolites, experimental results indicate the Mg-suite source region was not as rich in plagioclase-bearing rocks as previously estimated in model (I). Rather, experimental results suggest plagioclase undersaturated (low-Al) MgO-rich melts are parental to chromite-bearing troctolites, whereas PST is an indicator of magma-wallrock interactions (high-Al) within the lunar crust (model II) [3]. The process of (II) is likely restricted to the magma-wallrock interface due to slow diffusion rates of Al2O3 in basaltic melts [6] and thus, pink spinel-bearing lithologies are expected to represent a volumetrically minor component of the lunar crust [3]. The conclusions are supported by the paucity of PST among the Mg-suite samples and also the low number (~20-30) of global remote sensing detections of pink spinel [7, 8]. New estimates of the plagioclase undersaturated Mg-suite parent are provided by [3]. Lunar Crust: The predominance of intrusive Mg-suite samples is surprising considering that dense mare basalt flows cover ~18% of the lunar surface [9] and Mg-suite parental melts are 200 – 300kg/m3 less dense than mare basalts [10]. What prevented low-density Mg-suite parental melts from buoyantly erupting? Motivated by recent crustal density measurements from GRAIL [11], buoyancy-driven magmatic ascent of Mg-suite melts was investigated [12,13]. Results from [12,13] suggest present day, low-crustal densities measured by GRAIL (due to increased estimates of porosity) are needed to have prevented ancient, low-density Mg-suite parental melts from buoyantly erupting. Because the Mg-suite samples are ancient (> 4.1Ga), the results imply the primary lunar crust may have been fractured soon after solidification creating a porous, low-density barrier to Mg-suite eruptions [13]. The results also imply younger mare basaltic volcanism was driven by more than buoyancy-forces alone (e.g., hydrostatic forces may be required) [14]. New Mg-suite parental melt estimates by [3] are consistent with intrusive petrogenetic models, and potential regions of eruption are focused within the nearside southern highlands [13]. The Mg-suite eruptive area within the nearside southern highlands [13] is strongly correlated with remote detections of pink spinel (linked to Mg-suite) [5,7], and also positive Bouguer anomalies (potentially buried Mg-suite) [15]. The findings suggest the nearside southern highlands is the most promising region to explore ancient intrusions and possible volcanic deposits of the lunar highlands Mg-suite [13]. References: [1] Longhi, J. et al. (2010) GCA 74 784-798 [2] Warren, P.H. (1986) [3] Prissel, T.C. et al., (2016) Am Min 101, 1624-1635 [4] Elardo, S.M., et al., (2017) LPSC 48 #2450 [5] Prissel, T.C. et al., (2014) EPSL 403, 144-156 [6] Morgan, Z. et al. (2006) GCA, 70 3477-3491 [7] Pieters, C.M. et al. (2011) JGR 116(E6) [8] Sun, Y. et al., (2017) EPSL 465, 48-58 [9] Whitten, J.L. & Head, J.W. (2015) Icarus 247 150-171 [10] Prissel, T.C. et al. (2013) LPSC 44 #3041 [11] Wieczorek, M.A. et al. (2013) Science 339 671-675 [12] Prissel, T.C. et al. (2015) LPSC 46 #1158 [13] Prissel, T.C., et al., (2016) Icarus 277 319-329 [14] Head, J.W. & Wilson, L. (2017) Icarus 283 176-223 [15] Sori, M.M. et al., (2016) Icarus 273 284-295
Delivered As: 
Jennifer L. Whitten - Smithsonian Institution, MRC 315, PO Box 37012, Washington, DC 20013-7012, United States Stephen W. Parman - Department of Earth, Environmental & Planetary Sciences, Brown University, Providence, RI 02912, United States James W. Head - Department of Earth, Environmental & Planetary Sciences, Brown University, Providence, RI 02912, United States
SSERVI Identifier: