Global assessment of pure crystalline plagioclase across the Moon and implications for the evolution of the primary crust

Abstract

Recent advancements in visible to near infrared orbital measurements of the lunar surface have allowed the character and extent of the primary anorthositic crust to be studied at unprecedented spatial and spectral resolutions. Here we assess the lunar primary anorthositic crust in global context using a spectral parameter tool for Moon Mineralogy Mapper data to identify and map Fe-bearing crystalline plagioclase based on its diagnostic 1.25 μm absorption band. This allows plagioclase-dominated rocks, specifically anorthosites, to be unambiguously identified as well as distinguished from lithologies with minor to trace amounts of mafic minerals. Low spatial resolution global mosaics and high spatial resolution individual data strips covering more than 650 targeted craters were analyzed to identify and map the mineralogy of spectrally pure regions as small as~400 m in size. Spectrally, pure plagioclase is identified in approximately 450 targets located across the lunar surface. Diviner thermal infrared (TIR) data are analyzed for 37 of these nearly monomineralic regions in order to understand the compositional variability of plagioclase (An#) in these areas. The average An# for each spectrally pure region is estimated using new laboratory measurements of a well-characterized anorthite (An 96 ) sample. Diviner TIR results suggest that the plagioclase composition across the lunar highlands is relatively uniform, high in calcium content, and consistent with plagioclase compositions found in the ferroan anorthosites (An 94-98 ). Our results confirm that spectrally pure anorthosite is widely distributed across the lunar surface, and most exposures of the ancient anorthositic crust are concentrated in regions of thicker crust surrounding impact basins on the lunar nearside and farside. In addition, the scale of the impact basins and the global nature and distribution of pure plagioclase requires a coherent zone of anorthosite of similar composition in the lunar crust supporting its formation from a single differentiation event like a magma ocean. Our identifications of pure anorthosite combined with the GRAIL crustal thickness model suggest that pure anorthosite is currently observed at a range of crustal thickness values between 9 and 63 km and that the primary anorthositic crust must have been at least 30 km thick.

The formation of peak-ring basins: Working hypotheses and path forward in using observations to constrain models of impact-basin formation

“…This scenario also implies that peak rings should come from relatively shallow crustal levels, controlled by the geometry of the excavation and melting cavities. These predictions on stratigraphic uplift are consistent with the abundance of pure crystalline anorthosite in many of the peak rings of basins on the Moon (Hawke et al, 2003;Yamamoto et al, 2012;Cheek et al, 2013;Donaldson Hanna et al, 2014;Baker and Head, 2015). However, the hypothesized regions of stratigraphic uplift are distinct from the estimated stratigraphic uplift of central peaks in complex craters, which are commonly inferred to be constrained by the maximum depth of melting (Cintala and Grieve, 1998a,b;Tompkins and Pieters, 1999).…”

“…Peak rings on the Moon typically show outcrops of predominantly pure crystalline anorthosite and more noritic lithologies (Hawke et al, 2003;Whitten et al, 2011;Yamamoto et al, 2012;Cheek et al, 2013;Donaldson Hanna et al, 2014;Baker and Head, 2015). An exception is in regions of thinned crust such as the interior of the South Pole-Aitken basin.…”

“…Although there is still some uncertainty in the exact stratigraphy of the lunar crust, these mineralogic observations strongly indicate that peak rings are exposing crustal lithologies; mantle or very deep crustal materials are not exposed in the peaks rings except perhaps in areas of thinned crust. Furthermore, observations of large outcrops of pure crystalline anorthosite within the lunar peak rings at all sizes (e.g., Hawke et al, 2003;Cheek et al, 2013;Donaldson Hanna et al, 2014;Baker and Head, 2015) suggest that peak rings may be composed of materials uplifted from the upper crust (as opposed to the more noritic lower crust; Wieczorek and Phillips, 1997;Hawke et al, 2003). Alternatively, these results could imply an anorthositic crust to great depth (Yamamoto et al, 2012;Cheek et al, 2013;Donaldson Hanna et al, 2014), in which case lower crustal materials may also be exposed within the inner rings of impact basins (e.g., Kring et al, 2013;Potter et al, 2013a).…”

Mid‐Infrared Optical Constants of Labradorite, a Triclinic Plagioclase Mineral

“…In a telescopic study, Spudis et al (1984) suggested the presence of anorthosite, not by the presence of a 1.25-μm band but by the lack of other mafic absorption bands. In the past decade, high spectral and spatial resolution VNIR reflectance data sets from the SELENE Spectral Profiler (SP) and Multiband Imager (MI) and the Chandrayaan-1 Moon Mineralogy Mapper (M 3 ) have identified iron-bearing crystalline plagioclase with its characteristic 1.25-μm absorption band in diverse regions on the Moon (Cheek et al, 2013;Donaldson Hanna et al, 2014;Matsunaga et al, 2008;Ohtake et al, 2009;Pieters et al, 2009). Clementine ultraviolet-visible (UVVIS) imaging data sets examined the central peaks of over 100 impact craters across the Moon and demonstrated that the lunar crust is extremely anorthositic (Tompkins & Pieters, 1999).…”

“…Plagioclase consists a series of compositions from albite (NaAlSi 3 O 8 ) to anorthite (CaAl 2 Si 2 O 8 ) that indicate changing conditions during fractional crystallization. On the Moon, VNIR spectroscopy has been used to identify pure plagioclase using this feature only in areas that lack mafic phases (Cheek et al, 2013;Donaldson Hanna et al, 2014). This band, however, is easily masked by <15 vol.% of mafic phases, such as pyroxene or olivine (Crown & Pieters, 1987;Nash & Conel, 1974).…”