49th Lunar and Planetary Science Conference 2018 (LPI Contrib. No. 2083) 2553.pdf

CHARACTERIZING IMPACT MELT ON MERCURY. C. Leight1 and L. R. Ostrach2, 1Mount Holyoke Col- lege, South Hadley, MA, 01075, 2U.S. Geological Survey, Astrogeology Science Center, 2255 N. Gemini Dr., Flag- staff, AZ 86001. ([email protected])

Introduction: Orbital observations of the surface by the Mercury Surface, Space ENviroment, Geochemis- try, and Ranging (MESSENGER) spacecraft [1] have revealed the importance of volcanism over Mercury’s geologic history [e.g., 2]. Many of Mercury’s smooth plains regions have volcanic sources [e.g., 3,4], but

many smooth plains deposits have no clear origin. Spa- tially limited regions of smooth plains on Mercury typ- Xiao Zhao, 24 km D Fonteyn, 29 km D ically infill topographically low areas, such as craters, 10.58°N, 123.84°E 32.82°N, 95.51°E which further muddles determination of their origins and leads to the conclusion that some smooth plains could have an impact origin. Identification of impact melt on Mercury is difficult due to its similarity to other smooth plains deposits. To date, identification of impact melt on Mercury has typi- cally relied on three primary indicators: interior pond morphology (smooth texture, distinct crater-fill bound- Cunningham, 37 km D Degas, 54 km D 30.41°N, 157.0°E 37.08°N, 232.6°E aries, few superposed craters), crater-relations (i.e., a lack of embayed interior craters), and the presence of exterior deposits (i.e. exterior ponds and veneers) [e.g., 5-8]. Identifying these features generally require younger craters, where degradation has not yet heavily modified the surface. Images from the MESSENGER Mercury Dual Imaging System (MDIS) provide the means to further distinguish and characterize criteria for Unnamed 9.288, 47 km D Stieglitz, 100 km D the identification of impact melt on Mercury. 9°N, 288°E 72.53°N, 67.63°E Method: Using the MESSENGER Quickmap 8- color and monochrome maps [http://bit.ly/2wBAHJi], eight craters and one basin interpreted to be of relatively young age on the basis of superposed crater relations were selected [Figure 1]. Six of these craters displayed the characteristics traditionally used to distinguish im-

pact melt [e.g., 5, 6]; two craters and the basin displayed Eminescu, 129 km D Stravinsky, 129 km D embayment relations between superposed craters and 10.65°N, 114.2°E 51.9°N and 281.1°E floor materials, suggesting a volcanic origin. Monochrome mosaics of each crater were con- structed from MDIS Narrow Angle Camera (NAC) im- Figure 1. Craters and ages at different pixel scales and overlaid on the 166 basins used in this meters per pixel (mpp) basemap, obtained from the study. Images Planetary Data System (PDS). For each crater, images cropped from MDIS were selected on the basis of illumination conditions (to global monochrome emphasize morphology) and pixel scale (71 mpp to 6 Rachmaninoff, 305 km D mosaic. mpp). Images were processed in orthographic projection 27.66°N, 57.37°E using the USGS Projection on the Web [9], which is a The craters selected with material initially assumed public, online tool that allows registered users to process to be impact melt on the basis of interior pond morphol- PDS data using the Integrated Software for Imagers and ogy and the presence of distinct crater-fill boundaries by Spectrometers (ISIS3). Then, images were imported order of increasing size, were Xiao Zhao (24 km diam- into ESRI ArcGIS, where each crater was digitized to eter), Fonteyn (29 km diameter), Cunningham (37 km identify the geologic and any unique features within and diameter), Degas (54 km diameter), Stieglitz (100 km immediately surrounding the crater. diameter), and Eminescu (129 km diameter). The craters 49th Lunar and Planetary Science Conference 2018 (LPI Contrib. No. 2083) 2553.pdf

like hollows, were ignored as having no constrained re- lationship to both impacts and volcanism [e.g., 11]. Impact-Melt Features: Within the 6 craters with ma- terial interpreted to be impact melt, four common fea- tures were noted: (1) melt pools superposed on walls, (2) sub-kilometer width linear features [Figure 2], (3) flow features [Figure 3], and (4) low albedo features. All of the craters thought to contain impact melt material had melt ponds perched on wall terraces, and flows were Figure 2. Sub-kilometer linear features observed in also commonly observed on crater walls. The scale of Degas at 12 mpp. Image width ~13 km. the sub-kilometer width linear features seen in impact- melt ponds is vastly different than what is seen in the volcanically-embayed craters, occurring with widths not exceeding 1 km, and largely below 0.8 km, rather than on a scale of several kilometers. Flow features were visible from 25 mpp, but were easier to resolve at higher resolutions (e.g., 6 mpp), and so may be unresolvable in other craters with coarser image resolutions. Flow lobes and channels were found both between internal melt pools and at the edges of the crater floor. Xiao Zhao, Fonteyn, and Eminescu have low-albedo features in the Figure 3. Flow features observed on the interior pooled crater floors. crater wall of Degas. The criteria developed on the Moon for identifying impact melt are all still valid; interior pond morphology, initially assumed to have volcanic infill are, in increas- crater-relations, and the presence of exterior deposits ing size, Unnamed crater 9.288 (47 km diameter), Stra- are the most certain indicators of origin for smooth de- vinsky (129 km diameter), and Rachmaninoff basin posits. The unprecedented higher resolution of MDIS (305 km diameter). All of these craters are older than NAC images, however, allows for the development and those with deposits interpreted to be impact melt, with application of additional criteria for impact melt. These Rachmaninoff basin appearing the youngest due to the features are exclusive to the craters thought to contain dearth of superposed craters [10]. impact melt material. None of the morphological fea- The interior morphology and characteristics of each tures noted here are unique to this study or to Mercury, crater was digitized using MDIS NAC images com- and have been noted previously in the literature. How- prised of at least two resolutions and incidence angles ever, this study supports the idea that these features are as permitted by the available data. For each crater, the diagnostic of an impact origin for these smooth plains initial mapping was conducted at a coarser resolution deposits. and more detailed mapping was conducted at higher res- Further characterization of impact melt on Mercury olutions. Contacts between units in the crater (e.g., the will require the analysis of color data; the presence of crater wall and floor pond) were mapped. Common albedo features within the melt could be a definite indi- units identified were central peaks, blocky floor areas, cator of impact origin for melt deposits. Many features crater wall, floor melt pools, and interior and exterior are best viewed at 25 mpp or better resolution, and so melt ponds. For deposits interpreted to be impact melt, the analysis of more craters and their features at those interior features such as flow lobes or cracks were iden- resolutions would be beneficial. tified and mapped. Within both impact and volcanic Acknowledgements: This project was funded by the NSF melt deposits, cracks and tectonic features were identi- through grant number AST-1461200. Any use of trade, firm, or prod- fied and mapped as linear features, with causation at- uct names is for descriptive purposes only and does not imply endorse- ment by the U.S. Government. tributed later, if possible. Albedo features present across References: [1] Solomon, S.C. et al. (2001) PSS, 49, 1445-1465. all resolutions and incidence angles were also mapped. [2] Denevi, B.W. et al. (2009) Science, 324, 613-618. [3] Head, J.W. Results and Discussion: Several common morpho- et al. (2009) EPSL, 285, 227-242. [4] Murchie, S.L. et al. (2015) Ica- logical features were observed during mapping. Some rus, 254, 287-305. [5] Hawke, B.R., Head, J.W. 1977. In Impact and Explosion Cratering: Planetary and Terrestrial Implications: 851. [6] of the features, i.e. embayed craters and pooled floor Ostrach, L.R. et al. (2012) LPS 43, Abst. 1113. [7] Beach et al. (2012) materials, were used to identify the craters for this pro- LPS 43, Abst. 1335. [8] D’Incecco et al. (2015) PSS 119, 250-263. [9] ject and were expected. Others, such as the large-scale Hare, T.M. et al. (2015) 2nd Planetary Data Workshop, Abstract 7005. [10] Prockter, L.M. et al. (2010) Science, 329, 668-671. [11] linear features, were not expected. Anomalous features, Blewett, D.T. et al. (2011) Science, 333, 1856-1859.