HOW DO CRATER LAKES on MARS DEVELOP INLET VALLEYS? T. A. Goudge1, C

Home , Lakes on Mars, Stokes (Martian crater)

50th Lunar and Planetary Science Conference 2019 (LPI Contrib. No. 2132) 1223.pdf

HOW DO CRATER LAKES ON MARS DEVELOP INLET VALLEYS? T. A. Goudge1, C. I. Fassett2, and G. R. Osinski3, 1Jackson School of Geosciences, The University of Texas at Austin, Austin, TX, 2NASA Marshall Space Flight Center, Huntsville, AL, 3Centre for Planetary Science and Exploration/Dept. of Earth Sciences, University of Western Ontario, London, ON, Canada. (Contact: [email protected])

Introduction: Impact craters are ubiquitous on most In the final mechanism, if the topography upslope of planetary surfaces and provide enclosed topography that the crater rim is a closed basin, upstream fluvial activity is readily exploited by hydrologic activity to form lake would lead to ponding behind the natural dam formed by basins. Crater-hosted lake basins are observed on Earth the rim. Sufficient ponding could overtop the rim and [e.g., 1-3], and are abundant on Mars [e.g., 4-7]. A defin- erode through it as the temporary upstream lake drained ing geomorphic feature of impact craters is the rim, via a dam-breach flood [e.g., 12-14,18,19]. which sits elevated above the background terrain [8-10], typically by about ~1.5–2% of the crater diameter on Mars [11]. For tens-of-kilometer-scale craters, as is typi- cal for crater-hosted lakes on Mars [4-6], this represents relief on the order of hundreds of meters. The majority of crater-hosted paleolakes on Mars have incised inlet valleys that allowed flow of water into the basin. An obvious question thus arises: how was the topographic barrier of the crater rim originally breached by an inlet valley(s) to allow water to flow into the basin? This question is critical for understanding the formation and evolution of crater-hosted lakes on Mars (and Earth); however, it remains a poorly constrained problem. Past work on the Holden crater inlet valley indicates the inlet breach formed due to high-energy overflow flooding from a temporary lake in Uzboi Vallis [12-14]. In contrast, preliminary numerical modeling by [15] sug- gests that inlet valley breaching most likely occurs via ongoing impact cratering that works to remove the topographic barrier of the crater rim. This problem is also somewhat analogous to the classic geomorphology problem of forming transverse valleys that cut across moun- tain chains on Earth [e.g., 16,17]. Mechanisms for Inlet Valley Formation: Based on past work, our expectations for this problem, and analogy with formation of transverse valleys on Earth, we pro- pose four mechanisms for the development of crater lake inlet valleys (Fig. 1). These hypothesized mechanisms are an attempt at a framework within which to consider this problem, and we are entirely open to the possibility of additional, or combinations of, formation mechanisms. The first two mechanisms are similar, and involve re- moval of the topographic barrier of the crater rim through: (1) erosion and lowering of the crater rim at a rate higher than erosion of the surrounding terrain (Fig. 1a); or (2) deposition and raising of the exterior terrain to the elevation of the crater rim (Fig. 1b). In either case, a new, or previously existing, fluvial system would then be able to flow into the crater unimpeded [e.g., 15]. The third mechanism involves incision of the crater wall, initiated by surface runoff (from rainfall, snowmelt, or groundwater exfiltration). With ongoing runoff and in-

cision, retreat of the valley headwall could be sufficient Fig. 1. Schematic cartoons of a crater rim cross-section outlin- to establish a hydraulic connection with the exterior ter- ing four hypothesized mechanisms for inlet valley formation. rain, capturing any upstream drainage [e.g., 17]. 50th Lunar and Planetary Science Conference 2019 (LPI Contrib. No. 2132) 1223.pdf

Preliminary Study: We present preliminary results from a study to test between the four proposed mechanisms for a martian closed-basin lake contained in a ~73 km diameter impact crater (Fig. 2). We used two CTX [20] DEMs, produced using the NASA Ames Stereo Pipeline [21,22], and MOLA topography [23] to extract a longitudinal profile of the inlet valley (Fig. 2b), a profile across the inlet valley breach (Fig. 2c), and average topography from 10 radial profiles (Fig. 2d). From these data, we note three primary observations: (1) there is no evidence of high-standing crater rim topography near the breach (Fig. 2d; green dashed line is at expected crater rim position based on estimated crater diameter); (2) the inlet breach is not located at the lowest elevation location along the crater rim/margin (Fig. 2c); and (3) there is only one inlet valley, and no obvious evidence for dissection of the interior crater wall aside from this inlet. Of the four mechanisms outlined above, we suggest our observations are most consistent with formation via rim erosion and/or upstream infilling. Discussion: Several hundred crater-hosted paleolake basins exist on Mars, and they provide some of the best constraints on the activity of liquid water on the early martian surface [e.g., 4-7]. It has also been hypothesized that impact craters had a strong control on the fluvial evolution of early Mars by acting as a threshold and control for regional landscape integration [e.g., 6, 24-26]. A first step in forming a valley-fed, crater-hosted paleolake is establishing a hydraulic connection between the basin interior and exterior terrain, a process that remains poorly constrained. Here, and in planned future work, we aim to test competing hypotheses for how this process might have occurred, and whether there is a dominant mechanism(s) for this phenomenon across the planet. References: [1] Füchtbauer, H., et al. (1977), Geologica Ba- varia, 75:13. [2] Maloof, A., et al. (2010), GSAB, 122:109. [3] Arp, G., et al. (2013), GSAB, 125:1125. [4] Cabrol, N., and Grin, E. (1999), Icarus, 142:160. [5] Cabrol, N., and Grin, E. (2010), Lakes on Mars, Elsevier, pp. 1–29. [6] Fassett, C., and Head, J. (2008), Icarus, 198:37. [7] Goudge, T., et al. (2016), Geology, 44:419. [8] Pike, R. (1971), Icarus, 15:384. [9] Pike, R. (1980), Icarus, 43:1. [10] Melosh, H. (1989), Impact Crater- ing: A Geologic Process, Oxford Univ. Press, 245 pp. [11] Rob- bins, S., and Hynek, B. (2012), JGR, 117:E06001. [12] Pon- drelli, M., et al. (2005), JGR, 110:E04016. [13] Grant, J., et al. Fig. 2. Preliminary results for a closed-basin lake at −8.2°N, (2008), Geology, 36:195. [14] Grant, J., et al. (2011), Icarus, −159.4°E. (a) Overview of paleolake basin. Mapped lines show 212:110. [15] Enns, D., et al. (2010), LPSC 41, #2065. [16] inlet profile (red), breach profile (white), and rim radial pro- Hopkins, W. (1845), Trans. Geol. Soc. London, 7:1. [17] files (black). Image shows CTX DEMs B06_012008_1715- Stokes, M., and Mather, A. (2003), Geomorph., 50:59. [18] Ir- B06_011863_1714 and P02_001644_1713-P02_001842_1714 win, R., et al. (2002), Science, 296:2209. [19] Goudge, T., et al. and MOLA gridded topography overlain on the THEMIS [27] (2018), Geology, 47:7. [20] Malin, M., et al. (2007), JGR, global IR mosaic [28]. (b) Profile A–A’ along the inlet valley. 112:E05S04. [21] Shean et al. (2016), ISPRS J. Phot. Rem. Data extracted from CTX DEMs. (c) Profile B–B’ along the Sens., 116:101. [22] Beyer, R., et al. (2018), Earth Space Sci., crater rim/margin across the inlet breach. Data extracted from 5:537. [23] Smith, D., et al. (2001), JGR, 106:23,689. [24] Ir- CTX DEMs and MOLA. (d) Averaged radial profiles across the win, R., et al. (2005), JGR, 110:E12S15. [25] Irwin, R., et al. crater rim/margin (black = average, grey = ±2 standard devi- (2011), JGR, 116:E02005. [26] Barnhart, C., et al. (2009), JGR, ations). Data extracted from CTX DEMs and MOLA. Green 114:E01003. [27] Christensen, P., et al. (2004), Space Sci. Rev., dashed lines in (b) and (d) indicate expected location of crater 110:85. [28] Edwards, C., et al. (2011), JGR, 116:E10008. rim based on estimated crater radius.