ShadowCam: Shallower Small Craters in Lunar Permanently Shadowed Regions

SHADOWCAM: SHALLOWER SMALL CRATERS IN LUNAR PERMANENTLY SHADOWED REGIONS. Alexander J. Sonke1,2, Prasun Mahanti2, Brett W. Denevi3, Mark S. Robinson2, Jordan K. Ando4,5, Shuai Li5. 1Arizona State University, Tempe, AZ, USA, ([email protected]), 2Intuitive Machines, Phoenix, AZ, USA, 3Johns Hopkins University Applied Physics Laboratory, Laurel, MD, USA, 4University of Hawaiʻi at Mānoa, Honolulu, HI, 5Hawaiʻi Institute of Geophysics and Planetology, University of Hawaiʻi, Honolulu, HI. Introduction: Impact craters with diameters (D) <500 m are ubiquitous on the Moon, and their evolution (formation to erasure) produces and gardens the regolith and modifies surface topography in all geologic terrains [1,2]. Small craters are particularly useful for investigating terrains in unique environments, such as polar and permanently shadowed regions (PSRs), where crater shape can be used to infer regolith properties such as structure, stability, and volatile content. The ratio of crater depth to rim diameter (d/D) is of primary importance, such that fresh small craters typically form with 𝑑/𝐷 ≈ 0.2 [3, 4], and become shallower over geologic time from in-filling and degradation [5-7]. Previous studies have found that craters in diameter ranges of 100 m – 1km and 2.5 km – 15 km are shallower in polar regions than at lower latitudes, which was hypothesized to be due to the existence of subsurface volatiles [8, 9]. More recently, however, analysis of small craters within the Shoemaker crater PSR (150 – 400 m) found their d/D is comparable to non-PSR and non-polar crater populations [10]. Small craters in PSRs formed under low temperature and illumination extremes. PSRs do not receive direct solar illumination, thus observations of terrain in PSRs rely on diffuse secondary illumination scattered by nearby topographic highs [11]. We analyzed the morphometry of small craters (60 – 500 m) in PSRs from 2-m pixel scale images acquired by ShadowCam under secondary illumination [12]. Methods: We digitized small craters (D <500 m) on the floors of Faustini and Sverdrup craters by identifying them in orthorectified ShadowCam image mosaics (M017670865SE (170 cm/px) and M016815246SE (210 cm/px), respectively) and 6 m/px digital terrain models (DTMs) constructed from stereo observations [13]. To account for local topographic variations, we fitted a plane to the DTM pixels within 3 crater radii of each crater center and subtracted it from the DTM values. We then measured eight radial profiles from each crater center in 45º intervals. The height of the crater rim was computed by the sigma-rejection method of [14] to reduce the effects of overprinted craters or asymmetric degradation: rim elevations that fell below the mean of all eight rim points at a given crater were removed, and the rim height was taken as the mean elevation of the remaining points. Crater depth was then measured as the minimum of all profiles subtracted from the rim. We validated our morphometry algorithm at 50 random craters in our dataset by manually drawing profiles on the ShadowCam DTMs, finding the 𝑑/𝐷 from both methods in agreement (automated median = 0.041 vs. manual median = 0.042). Additionally, we verified our measurements at 47 craters that can be measured with at least 10 DTM pixels (200 – 500 m in diameter) in both ShadowCam DTMs and in the LOLA 80S 20-m DEM [15] using the methods of [16] (Fig.1). Craters on local slopes >5◦ or with D <10 DTM pixels were excluded from our analysis. Results: We extracted morphometric parameters of 455 small craters in the Faustini PSR and 728 small craters in the Sverdrup PSR. The 47 craters that can be reliably measured in both ShadowCam and LOLA DTMs had comparable d/D measurements between datasets (Fig.1). The ShadowCam DTMs generally captured greater crater depths, consistent with the sparse nature of LOLA data products. We observed no trend in the measurement comparison related to crater size in this diameter range (200 – 500 m), demonstrating that 10 DTM pixels adequately capture crater morphometry. We compared the overlapping diameter ranges (60 – 250 m) of our PSR-hosted craters with craters measured with LROC NAC data at the Apollo 16 Cayley Plains and Apollo 17 Taurus-Littrow Plains [16] (Figs. 2, 3). Craters with 𝑑/𝐷 ≤ 0.03 were removed from our dataset for this comparison to be consistent with [16]. We observe three distinctions between the PSR-hosted craters and the Apollo site craters. First, craters with d/D > 0.10 are nearly absent from the PSR groups. Second, the distribution of PSR-hosted craters is skewed such that their median d/D is 0.05, compared to 0.06-0.07 at the Apollo sites. Third, the depth-to-diameter relationship of the smallest craters (e.g. D <100m) has a steeper slope in PSRs than at the Apollo sites, consistent with a higher density of shallow craters (in the d/D = 0.03 – 0.04 range) in PSRs. Fig 1. Depth/diameter of small craters (200 – 500 m) in the Faustini and Sverdrup PSRs as measured in LOLA (20 m/px) versus ShadowCam (6 m/px). Discussion: Our morphometric analysis of Faustini and Sverdrup small craters show that they are shallower than their non-PSR counterparts (median d/D =-0.05 vs. -0.06-0.07, respectively, at D = 60 – 250 m and 𝑑/𝐷 > 0.03). We also observe that there are fewer craters with d/D > 0.1 in these PSRs, and more with d/D < 0.04, especially at D < 100 m. While small crater depths in Shoemaker PSR (150 – 400 m) were not interpreted as significantly different from non-polar and non-PSR crater depths [10], a threshold shallowing diameter below 150 m would allow for consistency between that interpretation and the results presented here. The shallow crater depths measured here are likely caused by regolith properties, which may or may not include the influence of volatiles. Regolith/ice mixtures limited to the top few meters of regolith could explain the shallower depths observed at smaller diameter ranges. It is also possible that porosity or temperature-dependent regolith properties result in initially shallower small crater formation in PSRs than non-PSR areas, or in enhanced degradation, which would be consistent with the dearth of craters with d/D > 0.1 and diameters 10 – 80 m observed here and in [8, 17]. We note from our ShadowCam observations of equatorial PSR analogs that secondary illumination does not increase the visibility of shallow small craters. Rather, shadows in crater interiors are diffuse when under secondary illumination compared to primary illumination. Nevertheless, the shallow craters we measured are unambiguously identifiable under secondary illumination in ShadowCam images, including those with d/D < 0.04, which, elsewhere, are considered unrecognizable [18]. In addition, our preliminary assessments of shaded relief maps produced from ShadowCam DTMs found no obvious differences in crater morphology when compared to LROC NAC DTMs. While the overall significance of the effects of secondary illumination are still under investigation, we emphasize that it is possible that such shallow small craters exist elsewhere on the Moon, though it is unlikely that their increased detection in PSRs is strictly related to lighting conditions. Conclusion: We observe that the d/D of small craters in PSRs are lesser than those of crater populations found in equatorial regions, likely due to different regolith properties, and possibly related to ice content. Small craters may form with initially shallower d/D in PSRs, or they may experience enhanced degradation, either by surface modification immediately after formation, or over the lifetime of the crater, or some combination thereof. References: [1] Robinson, M. S. et al. (2015) Icarus, 252, 229-235. [2] Speyerer, E. J. et al. (2016) Nature, 538.7624, 215-218. [3] Pike, R. (1974) GRL, 1.7, 291-294. [4] Basilevskii, A. T. (1976) LPSC Proc. 1, 1005-1020. [5] Ross, H. P. (1968) JGR, 73.4, 13431354. [6] Craddock, R. A. and Howard, A. D. (2000) JGR: Planets, 105.E8, 20,387-20,401. [7] Fassett, C. I. et al. (2022) JGR: Planets, 127.12, e2022JE007510. [8] Kokhanov, A. A. et al. (2015) Sol. Syst. Res., 49.5, 295-302. [9] Rubanenko, L. et al. (2019) Nat. Geosci., 12.8, 597-601. [10] Basilevsky, A. T. and Li, Y. (2024) PSS, 241, 105839. [11] Mahanti, P. et al. (2023) J. Astr. Space. Sci, 40.4, 131-148. [12] Robinson, M. S. et al. (2023) J. Astr. Space. Sci, 40.4, 149171. [13] Manheim, M. R. et al. (2024) 55th LPSC, #3040. [14] Robbins, S. J. et al. (2017) Meteoritics & Planet. Sci., 53(4), 583-637. [15] Barker, M. K. et al. (2023) PSJ, 4, 183. [16] Mahanti, P. et al. (2018) Icarus, 299, 475-501. [17] Denevi, B. W. (2024) 55th LPSC, #1491. [18] Basilevsky, A. T. et al. (2014) PSS., 92, 77-87. Fig 3. Probability density functions of small crater populations (60 – 250 m) in the Sverdrup and Faustini PSRs versus Apollo 16 and 17 plains, after removing d/D <= 0.03. Fig 2. Depth versus diameter of small craters (60 – 250 m) in Sverdrup and Faustini PSRs and the Apollo 16 and 17 plains. Dashed lines are constant depth/diameter ratio values. PSR-hosted craters with d/D <= 0.03 not shown.