The Formation of the Schrödinger Basin: 3D Impact Simulations in a Heterogeneous Target

THE FORMATION OF THE SCHRÖDINGER BASIN: 3D IMPACT SIMULATIONS IN A HETEROGENEOUS TARGET. D. P. Kallenborn1, G. S. Collins1, T. M. Davison1, D. A. Kring2, 1Earth Science and Engineering, Imperial College London, UK ([email protected]), 2Lunar and Planetary Institute (USRA), Houston, TX. Introduction: The Schrödinger basin is a ~320 km diameter impact crater on the farside of the Moon, close to the lunar south pole. The complex crater is the best-preserved basin of its size and has an impressive peak-ring structure [1]. The Schrödinger basin plays a key role in lunar geologic history. It formed around 3.8 billion years ago at the western rim of the pre-existing South Pole-Aitken basin, the largest and oldest basin on the Moon (Fig. 1). Some of the oldest materials, possibly even primordial crust, might have been excavated and could be exposed in the peak ring [2, 3]. The Schrödinger basin plays a prominent role in the upcoming Artemis missions, Farside Seismic Suite and the proposed Endurance mission. A detailed understanding of the basin formation process and the distribution of materials will be essential for planning those missions and interpreting collected data. Figure 1. Schrödinger basin (S, red circle) in Azimuthal Equidistant projection of Global Morphology [14] (left) and GRAIL crustal thickness [11] (right), centred on the basin centre. White dashed lines represent the western rim of the SPA (best-fit topographic and outer topographic ellipse) based on [15]. The white dot marks the lunar south pole. Profile lines (500 km) extend from west to east and south to north through the basin centre (black, dashed); see Fig. 2. The Schrödinger basin was formed in a complex heterogenous target with laterally varying pre-impact topography and crustal thickness, caused in part by several pre-existing basins, most prominently SPA. In the south, Schrödinger overlaps the pre-existing Amundsen-Ganswindt basin (~378 km) [4] which introduces further topographic and sub-surface complexities. Asymmetric ejecta distribution indicates an oblique impact, potentially on a SSE to NNW trajectory [5]. Several asymmetries within the Schrödinger basin—including its peak ring and crater wall—are evidence of this complex interplay of target heterogeneity and impact angle and azimuth [1, 2, 6, 7]. This work aims to investigate the role each factor played in the basin formation process by systematically analysing the effects of pre-existing topography, crustal thickness, and impactor trajectory in three-dimensional numerical models. Methods: We use the iSALE3D shock physics code [8, 9] to numerically model the formation of the Schrödinger basin. We simulate a range of oblique impact angles (20°, 30°, and 45°) for three impact bearings: 33.5° west of north and 22.3° east of north based on recent work on the ejecta distribution around Schrödinger [5] and a south-to-north trajectory for comparison. To determine a best-fit scenario, we compare our simulations to LOLA Topography [10] and GRAIL Crustal Thickness data [11]. SPA-dominated layer setup. We initially focus on the dominating influences of SPA which causes an eastward sloping topography and crustal thinning from 40 to 20 km (Fig. 1) across the pre-impact terrain. In our initial models we select parameters based on successful 2D simulations of the basin [3], where a 25 km asteroid impacts a granitic crust and dunite mantle with a vertical velocity of 15 km/s. For computational expediency we use the same material model to represent the impactor and crust, implying an impactor density of 2650 kg/m3. We use the “block model” for acoustic fluidization to facilitate late-stage collapse [12]. The resolution of the simulations is 1250 m (10 cells per projectile radius, CPPR). Including pre-existing craters. In a second step, we add complexity to the three-dimensional layer setup by considering the influence of the pre-existing Amundsen-Ganswindt basin on the formation of Schrödinger. We assume an almost pristine condition of Amundsen-Ganswindt at the time of the Schrödinger impact (~3.8 Ga). We adjust the topography and crust-mantle interface in the simulation setup by creating a synthetic Amundsen-Ganswindt basin based on a scaled, average Schrödinger profile. Our initial models have shown that the 2D simulation parameters do not perfectly translate to the 3D simulations because of the oblique impact scenario (see below). We account for observed discrepancies in basin size and central uplift shape by iteratively adjusting size and speed of the impactor, as well as the acoustic fluidization parameters. To resolve the impactor diameter (22 km) in the 45° scenario presented below, we reduce the cell size to 1 km (i.e., 11 CPPR). Results and Discussion: SPA-dominated layer setup. Initial simulated basin diameters range from 363 (33.5° W of N, 45°) to 381 km (S to N, 20°) overestimating the final basin size by 10 to 20%. In all three impact scenarios, basin sizes increase with shallower impact angle, suggesting a more complex scaling relationship between impactor diameter and velocity and final crater size. Geometric crater centres are offset from the point of impact as expected for oblique impacts [13]. The assumed eastward sloping pre-impact topography causes asymmetry in the simulated crater rim, with higher crater walls in the west, consistent with observed topography (Fig. 2). Figure 2. Surface topography and crust-mantle interface along the west to east (top) and south to north (bottom) profiles for a 45° impact angle and the three different bearings (left, middle and right). LOLA topography and GRAIL data (black) with their associated uncertainties (grey). For simulated profiles the origin (0,0) is defined as the simulated crater centre and profiles are offset accordingly. S-N transects (Fig. 2) that capture the main component for all three impact bearings show an offset between the simulated geometric crater centres and the centres of the structural uplifts. Additionally, the S-N profiles show a wider and deeper annular crustal bulge in the uprange direction (south). These findings suggest that for oblique impacts the position and shape of the of the central uplift may be used to constrain the impact bearing [13]. Including pre-existing craters. To form the Schrödinger basin in the right location and accounting for the observed offset between point of impact and crater centre in the initial simulations, the impactor hits the surface close to or on top of the northern AmundsenGanswindt crater rim. The modified impactor parameters produce a 329 km basin for a 33.5° W of N bearing and 45° impact angle (Fig. 3). Ejecta distribution will vary depending on impact bearing and angle. For the presented impact scenario Schrödinger ejecta covers most of the AmundsenGanswindt basin (~1 km thickness). The AG basin’s peak ring was likely similar in height to Schrödinger’s (1–2.5 km) [3] but is not preserved at the surface. Schrödinger ejecta, in combination with other, smaller impact events that have degraded the AG basin, could have hidden the putative AG peak ring. The simulated S-N crust-mantle interface now shows a discernible second peak underneath the southern rim which is evidence of the pre-existing Amundsen-Ganswindt basin (Fig. 3). The pre-impact topographic low in AG’s basin interior causes an almost absent Schrödinger crater rim in the south which is consistent with observed LOLA topography. Figure 3. Plan view of surface topography (left) with simulated Schrödinger basin (black, dashed) on top of synthetic Amundsen-Ganswindt basin (blue, dashed). Surface topography and crust-mantle interface (right) for a 33.5° W of N and 45° impact scenario (red) compared to LOLA topography and GRAIL crustal thickness (black). A distinct peak ring is not visible in the simulated surface topography due to the chosen cell sizes which are insufficient to resolve the topography of Schrödinger’s peak ring. Conclusions and Future Work: The simulations successfully explain some of the topographic asymmetries in the Schrödinger basin. Including the Amundsen-Ganswindt basin in the simulation setup produces a significantly better match to the geophysical data. This shows that the added complexity that can be facilitated with 3D models enhances the accuracy of numerical simulations. Findings of this work will therefore not only improve our understanding of the formation the Schrödinger basin, a crater with high scientific relevance, but will also provide a guideline for numerical modelling of crater formation in other complex and heterogeneous settings. Future work will include a comprehensive analysis of the interaction of impact trajectory and lateral variations in crustal thickness as well as an investigation into the remaining discrepancies between geophysical data and simulation results. Acknowledgments: We gratefully acknowledge the developers of iSALE. We thank Caroline Chalumeau whose MSci project provided the inspiration for the work presented here. References: [1] Shoemaker E. M. et al. (1994) Science, 266, 1851–1854. [2] Kramer G. Y. et al. (2013) Icarus, 223(1), 131–148. [3] Kring D. A. et al. (2016) Nature Comm., 7(1), 13161. [4] Head III, J. et al. (2010) Science, 329(5998), 1504–1507. [5] Kring, D. A. et al. (2025) Nature Comm., 16(1), 1–7. [6] Kring, D. A. et al. (2017) GSA Today. [7] Gulick S. P. S. et al. (2024) LPSC, 55, Abstract #2234. [8] Elbeshausen D. et al. (2009) Icarus, 204(2), 716–731. [9] Elbeshausen D. and Wünnemann K. (2011) Proc. 11th Hypervel. Impact Symp., Vol. 4, 287–301. [10] Smith D. et al. (2010). Geophys. Res. Letters, 37(18). [11] Wieczorek M. A. et al. (2013) Science, 339(6120), 671–675. [12] Wünnemann K. and Ivanov B. A. (2003). Planet. Space Sci., 51, 831–845. [13] Collins G. S. et al. (2020) Nature Comm., 11(1), 1480. [14] Planetary Data Systems (2018) Lunar Reconnaissance Orbiter. [15] Garrick-Bethell I. and Zuber M. T. (2009) Icarus, 204(2), 399– 408.