GRAIL Gravity Signatures of Basalts in Mare Imbrium

GRAIL GRAVITY SIGNATURES OF BASALTS IN MARE IMBRIUM J. N. Levin1 , A.J Evans1 , J.W. Head III 1 and J. T. Keane2, 1Department of Earth, Environmental, and Planetary Sciences, Brown University, RI, USA ([email protected]), 2 Jet Propulsion Laboratory, California Institute of Technology, CA, USA Introduction: The dark basaltic lunar plains (maria) that cover much of the lunar nearside inform about the magmatic history of the Moon. In this work, we focus on Mare Imbrium: the unit that fills and surrounds the Imbrium giant impact basin. The mare basalts that fill and surround the Imbrium basin as part of the Mare Imbrium unit are thought to be among the youngest and thickest on the lunar surface [1-4]. The relative scarcity of impact craters suggests that parts of Mare Imbrium are as young as 2 Gya [1], and also complicates efforts to measure mare thickness using morphological inferences. Consequently, the thickness, and volume of Mare Imbrium are poorly constrained even compared to other mare units, limiting our understanding of the thermal state of the nearside after the Imbrium impact. Gravity measurements from the Gravity Recovery and Interior Laboratory (GRAIL)[5] mission have been instrumental in improving our measurements of thickness and total volume of lunar maria [6,7]. However, because of the inherent non-uniqueness of gravity, GRAIL-based measurements of mare volume are often dependent on simplifying assumptions about the structure, composition, and compensation state of each region. However, because of the size and youth of the Imbrium impact basin, and its location within the Procellarum KREEP terrane (a region that hosts the majority of heat-producing elements on the lunar surface), the present-day state of the subsurface is poorly constrained in this region. In this work, we use spatio-spectral localization of GRAIL gravity to bound the distribution of mare mass within Mare Imbrium, and benchmark against morphological and spectroscopic constraints from remote-sensing observations. We also compare results with isostasy-based mare-thickness predictions to better constrain our understanding of the compensation state in the region before the emplacement of mare. Constraining Mass in Mare Imbrium: Following previous work [e.g. 6,8], we localize Free-Air GRAIL gravity (model GRGM1200B [9]) and Lunar Orbiter Laser Altimeter [10] topography using a taper with a 5º radius spatial window and a bandwidth of L = 50. We then calculate the admittance as the ratio of the spectral cross-power of gravity and topography to the spectral power of topography. The same computation is performed on a suite of synthetic free-air gravity fields that are generated using finite-amplitude topography. We compute synthetic gravity fields for a structure with a constantdensity and constant-thickness layer of mare over a constant-density crust. We vary three parameters: mare thickness, mare density, and crustal density to and compute synthetic localized admittance for each set of parameters. Synthetic admittance is then compared to to localized admittance from GRAIL by computing the root-mean-square (RMS) error between degrees 300-500. Models with RMS error within 15% of the minimum error are used to identify the most likely parameter combinations to explain GRAIL gravity observations. Finally, mare-mass-per-unit-area is calculated as the product of mare density and mare thickness to investigate the parameter space while accounting for the inverse relationship between thickness and density. We show the mean, minimum, and maximum values of mare-mass per-unit area calculated for windows centered on a 2.5º spaced grid over Mare Imbrium (Fig. 1a-c). The mare-thickness and mare-density values that correspond with the minimum and maximum of our mass calculations are shown in panels e,f,h, and i of Figure 1. Comparison to observables: Our results show an increase in top-layer mare-mass in the western portion of Mare Imbrium. The increase in mass corresponds with the location of the youngest mare units in the region [1] (Fig 1d). Per observations from Clementine, Lunar Prospector, and Moon Mineralogy Mapper (M3), the western basalts are titanium-rich and ironrich relative to older units [e.g. 11-13]. Consequently, younger basalts are expected to be denser, which is consistent with our model results. By incorporating morphological and spectral observations with our modeling, we can infer the volumetric properties of particular units from otherwisenon-unique gravity solutions. For example, one of the youngest crater-counted units in Mare Imbrium (~2 Gyr [1]), located at ~25ºN 335ºE (Fig. 1d), corresponds to morphologic estimates of mare thickness > 1.5 km, and observations from nearby Lambert crater bound the maximum thickness of the unit to 4.1 km [3,4]. Applying these thickness constraints to our model results, we find that the density of the unit at 25ºN 335ºE must exceed 3600 kg/m3. Our current model assumes that the density of the material underneath the top-most unit is similar to the highland crust, and thus is not allowed to exceed 3000 kg/m3. However, this is inconsistent with morphological observations of younger mare units covering older basalts. To account for this, future modelling will allow for higher densities in the bottom unit. Comparison to other GRAIL-based mare thickness measurements: Recent work [7] provided mare thickness estimates under the assumption of premare isostasy (Fig. 1g). Our results support their predictions of > 8 km of mare material in the center of the Imbrium impact basin, which is also in agreement with morphologic predictions [2]. However, our results differ in the lateral extent of the thickest part of this unit. The most striking difference is the discrepancy in thickness in the North-East corner of Mare-Imbrium, where the isostatic model predicts thickness of ~6km (Fig. 1g), and our model predicts thickness of 500 m or less in accordance with morphological inferrences [3]. This difference suggests that in the north-eastern portion of the Imbrium basin some of the and the gravitational signature must be attributed to subsurface loading due to mantle uplift during basin formation or later intrusive activity, and further suggests that the Imbrium basin did not reach an isostatic state prior to being filled with mare basalts. Summary: We detect young, dense mare basalts in the western portion of Mare Imbrium using GRAIL gravity, consistent with morphological and spectroscopic observations of the area. References: [1] Hiesinger, H. et al. (2011) Recent Advances and Current Research Issues in Lunar Stratigraphy, GSA, 477. [2] Head, J.W. III (1982) The Moon and the Planets, 26, 61-88 [3] Dehon, R.A. (1979) NASA-CR-158784 [4] Thomson, B. J. et al. (2009) GRL, 36 [5] Zuber, M. T. et al. (2013) Science, 178, 3-24 [6] Gong, S. et al. (2016) JGR: Planets, 121, 854-870 [7] Broquet, A. and Andrews-Hanna, J. C. (2024) Icarus, 408, 115846 [8] Wieczorek, M. A. et al. (2013) Science, 339, 671-675 [9] Goossens, S. et al. (2020) JGR: Planets, 125 [10] Smith, D. E. et al. (2010) Space Sci Review, 150, 209-241 [11] Staid, M. I. et al. (2011) JGR: Planets, 116. [12] Elphic, R. C. et al. (2002) JGR: Planets, 107, 8–1-8–9 [13] Thiessen, F. et al. (2014) Planetary and Space Science, 104, 244-252. [14] Robinson, M. S. et al. (2010) Space Sci Review, 150, 81-124. Figure 1. (a) Mean (b) Minimum and (c) Maximum of mass-per-unit-area of lowest-error models. We plot corresponding density (e-f) and thickness (h-i) values for the minimum (e, h) and maximum (f,i) cases. Panel (d) shows crater size-frequency model ages [1] over a morphologic basemap [14]; the example location analyzed in this work is marked with a filled polygon. Panel (g) shows mare thickness inferred using a pre-mare isostatic assumption [7]. The region of discrepancy between our model is circled in red.