Compositional variations of the lunar crust: Results from radiative transfer modeling of central peak spectra

Abstract

[1] We present model mineralogy of impact crater central peaks combined with crustal thickness and crater central peak depth of origin models to report multiple perspectives of lunar crustal composition with depth. Here we report the analyses of 55 impact crater central peaks and how their compositions directly relate to the lunar highlands sample suite. A radiative transfer model is used to analyze Clementine visible plus near-infrared spectra to place compositional constraints on these central peak materials. Central peaks analyzed are dominantly magnesian-and plagioclase-poor; strong compositional similarities to lunar Mg-suite materials are evident. Relative to crustal thickness estimates, central peak mineralogy becomes more plagioclase-rich as the crust thickens. Relative to the crust-mantle boundary, the origin of peaks with dominantly mafic mineralogy are confined to the lower crust and primarily within the South-Pole Aitken and Procellarum KREEP Terranes (PKT); additionally, central peaks with anorthositic mineralogy (>60 vol % plagioclase) are transported to the surface from all depths in the crustal column and confined to the Feldspathic Highlands Terrane (FHT). The discovery of mafic and magnesian materials, consistent with Mg-suite rocks of the sample collection, in all lunar terranes suggests that the process and sources that give rise to these types of rocks is not unique to the PKT and not necessarily dependent on incompatible elements for formation. The identification of ferroan and magnesian anorthositic material near the crust-mantle boundary of the FHT is also inconsistent with an increasing mafic/feldspar ratio and Mg' with depth in the crust.

Compositional heterogeneity of central peaks within the South Pole‐Aitken Basin

“…On average, the crust is~40 km thick, as shown by recent models incorporating Gravity Recovery and Interior Laboratory (GRAIL) gravity and Lunar Orbiter Laser Altimeter (LOLA) topography data [Wieczorek et al, 2013]. Several lines of evidence from lunar sample analyses as well as remote sensing studies suggest that the mafic content of the crust increases with depth [Reid et al, 1977;Ryder and Wood, 1977;Bussey and Spudis, 2000;Tompkins and Pieters, 1999;Cahill et al, 2009;Ohtake et al, 2009]. However, impacts throughout lunar history have fractured, brecciated, mixed, melted, and redistributed crustal materials.…”

“…This reasoning sets the maximum depth of melting as the minimum depth of origin of central peak material. The maximum depth of melting is estimated to be approximately 0.109D 1.08 , where D is the final crater diameter [Cintala and Grieve, 1998;Cahill et al, 2009]. In central peak craters, the deepest material exposed is located in the central peak.…”

“…Central peak craters are especially useful, as they uplift material from beneath the excavation cavity, often from depths of~10 km or greater (depending on crater diameter) [e.g., Melosh, 1989;Cintala and Grieve, 1998]. A number of studies have been performed using central peak craters to evaluate the stratigraphy of the lunar subsurface [e.g., Tompkins and Pieters, 1999;Wieczorek and Zuber, 2001;Cahill et al, 2009;Nakamura et al, 2009;Song et al, 2013]. The primary objective of these papers has been to discern global and regional stratigraphic trends through the compositional assessment of multiple central peaks.…”

Obvious problems in lunar petrogenesis and new perspectives

“…With all the observations considered, a revised assumption is needed, assumption 3*: Regolith compositions everywhere are dominated by immediately underlying bedrock, with both lateral and vertical mixing possible but limited (Cahill et al, 2009). …”

“…Impact crater central peaks offer a unique opportunity to remotely examine material that was once deep in the crust (Cahill et al, 2009). The majority of peaks analyzed had compositions similar to Mg-suite rocks of the lunar sample collection, independent of lunar terrain, contrary to one of the principal conclusions by Korotev (2005) based on lunar highland meteorites.…”