What Lies Beneath: What Largest Ejected Boulders Tell us About the Substrate

WHAT LIES BENEATH: WHAT LARGEST EJECTED BOULDERS TELL US ABOUT THE SUBSTRATE. W. K. Carter1 and M. S. Robinson2, A. K. Boyd2, R. V. Wagner2, J. S. Barnett2. 1Arizona State University SESE, Tempe AZ 85287 (wkca[email protected]), 2Intuitive Machines 101 E. Jackson St. Phoenix AZ 85004. Introduction: Since January 2023, ShadowCam [1] has provided high-resolution images of lunar permanently shadowed regions (PSRs). PSRs are thought to sequester volatiles due to the cold surface and subsurface temperatures (<110K) [2,3,4]. As volatiles vary in the subsurface, the strength of the substrate varies as well [5]. Differences in strength result in different maximum ejected block size among similarly sized craters during crater formation with higher subsurface competency resulting in a larger maximum ejected block size [6, 7]. In 1971, H. J. Moore [6] measured over 100 crater and boulder diameters from Lunar Orbiters 2, 3, and 5 photographs, images acquired from the Apollo 10 Command Module, and direct measurements of terrestrial craters. Moore proposed the following relationship between the largest ejected boulder (B) and crater diameter (D): B=KD2⁄3 B is boulder diameter (cm), D is crater diameter (cm), and K is a constant ranging from 0.5 to 1.5 [6]. Moore used measurements from craters (6 cm to 85 km diameter) across terrestrial (natural and anthropogenic) and lunar regions. [8] used this relation and found that the value K decreases as craters age. Here, we test the hypothesis that K varies amongst seven units, including five equatorial regions: Moore’s data set (MooreLunar), Cayley plains (Light Plains), Mare Tranquillitatis (Old Mare), units P57 and P58 from [9] (Young Mare), and equatorial analog craters and two south pole regions: PSRs and temporarily shadowed regions. Methods: We derive an equatorial control dataset from the four units to investigate differences in K across the two polar units. To ensure the equatorial control group is comprised of similar regions, we use one-way analysis of variance (ANOVA) to test if the K values of the regions are significantly different (P<0.1) and eliminate those regions from the control group. Measurement collection for the equatorial terrains was a two-step process. First, from small incidence angle (<40°) NAC images, we identified fresh craters 100 m to 1000 m in diameter by their characteristic high-reflectance ejecta deposits. Second, we examined larger-incidence angle NAC images of the same craters to measure crater diameter and the diameter of the largest ejected boulder (Fig. 1). For the PSRs, the large effective incidence angle and diffuse lighting lowers the contrast between the ejecta of fresh craters and substrate. Therefore, within ShadowCam images, we use the morphologic criteria of sharp rims and the existence of blocks to identify fresh craters and use the same ShadowCam images to measure crater and block sizes. We then perform a one-way ANOVA of the control group against each of the polar regions to assess if the populations are statistically significantly different. Figure 1: Example of fresh rocky crater and largest ejecta boulder in PSR SP_887280_1688800 [M066003083S] Results: In the equatorial regions, the Young Mare K is found to be significantly different from the other equatorial areas (P=0.00007; Fig. 2, 3). For this reason, the Young Mare is excluded from the equatorial control dataset. K value differences for the other equatorial units are not statistically significant (P=0.72), with the result that our equatorial control group is a compilation of MooreLunar, PSR Analog, Old Mare, and Light Plains. The K of the Temporary Shadowed Regions is also significantly different from the control group (P=0.0001; Fig. 2, 3). The mean K value of craters within PSRs is not significantly different from the control group (P=0.23; Fig. 2, 3). Conclusions: We verify Moore’s boulder and crater relationship with a larger dataset; K falls within 0.5-1.5 for lunar terrains (Table 1). We show that 4 of the 5 equatorial units investigated are indistinguishable (PSR analog, Old Mare, Light Plains, and the original Moore dataset), and use those combined terrains as our control group. The difference between the equatorial Young Mare unit and other equatorial terrain is statistically significant (Table 1). We hypothesize that this difference indicates a stronger substrate causing more competent rock to be excavated during crater formation. Fresh craters in the South Pole Temporary Shadowed region significantly differ from the equatorial Control group. The South Pole Temporary shadowed unit has lower K values and block sizes (Table 1). We hypothesize that the smaller boulder sizes in this region are due to the unique thermal environment with more frequent and more intense temperature cycling as boulders pass in and out of shadow. This intense thermal cycling results in shorter block lifespans. Alternatively, these transiently illuminated areas may harbor subsurface volatiles that alter the regolith properties such that ejected boulder populations exhibit smaller diameters relative to the other units. When compared with the control group, craters within PSRs did not have a significant statistical difference, indicating that the subsurface in PSRs is similar to equatorial areas, and if present, frozen volatiles did not influence the K relationship. [1] Robinson M.S. et al. (2023) JASS, 40(4), 149171. [2] Paige D.A. et al. , (2010) Science, 330, 479482. [3] Lucey et al. (2021) Acta Astronautica, 180 25–34. [4] Brown et al. (2022) Icarus, 377, 1114874. [5] Moon, S., et al. (2021) GRL, 48, e2020GL090780. [6] NASA Manned Spacecraft Center (1971) NASA SP 232, 226 pp. [7] Cintala, M. et al. (1995) NASA TM 104804. [8] Watkins, R. N. et al. (2019) JGR, 124, 2754-2771. [9] Hiesinger, H., et al. (2000) JGR, 105(E12), 29239–29275. Acknowledgments: The NASA ShadowCam investigation supported this work, all ShadowCam data used in this work are archived in the NASA PDS. Table 1: K, 95% prediction bounds (Kmin Kmax), crater statistics, and median boulder diameter for the terrains studied. Figure 2: Box-plot comparison of estimated K values for all terrains. Red lines are medians, notches are 95% median prediction bounds, blue boxes are the 1st and 3rd quartiles (50% of the data is within the blue box), whiskers are the 99.3% estimators, and red pluses are individual data points beyond the whiskers (outliers). Figure 3: Comparison of the groups using Fisher’s Least Significant Difference Procedure to test P < 0.1. The difference in mean of K for Young Mare and South Pole Temporary Shadow are statistically significant. The other 6 groups have a K mean that is not significantly different from the control.