THE BOSUMTWI IMPACT CRATER, A TERRESTRIAL RAMPART CRATER. G. Wulf1 and T. Kenkmann1, 1Institute of Earth and Environmental Sciences -Geology-, Albert-Ludwigs-University Freiburg, Ger-many, gerwin.wulf@geologie.uni-freiburg.de.

Introduction: The Bosumtwi impact crater in

Ghana (06°30`N, 01°25`W) is one of the youngest and best preserved terrestrial impact craters [1, 2]. The 1.07 Ma old impact structure has a pronounced crater rim with a diameter of 10.5 km [3]. The circular struc-ture is somewhat distorted in the SE sector by the northern flank of the NE–SW trending Obuom Moun-tain Range [4] (Fig. 1a). The impact structure forms a hydrologically closed basin that is almost completely filled by the 8.5 km diameter Lake Bosumtwi. The impact crater was intensely studied in the framework of various drillings campaigns, e.g., the International Continental Scientific Drilling Program (ICDP) drill-ing in 2004 [5, 6]. Suevites (polymict, glass-bearing breccias) and polymict clastic impact breccias were found in the boreholes as well as in the northern and southern parts within the crater structure. Although chemical weathering is intense in the tropical rain for-est environment and led to the formation of locally thick lateritic soils, a lot of breccias associated with the formation of the crater are still preserved outside of the crater rim (see review by [2]). Moreover, polymict and monomict clastic breccias were reported from numer-ous places around the impact crater whose impact origin was not unambiguously confirmed [2]. A strik-ing feature of the Bosumtwi impact structure is a shal-low, near-circular, depression at 7-8.5 km from the crater center directly beyond the crater rim, followed by a shallow outer topographic ring feature at 18-20 km radial distance from the impact center [7, 8] (Fig. 1b). This feature is visible in radar satellite images as well as in aero-radiometry data, indicating lithological as well as topographic control [4, 8, 9]. It was suggest-ed that preferential removal of ejecta within the area just outside of the crater rim could be the reason for this shallow depression [8] or original depositional patterns as well as impact-induced concentric fractur-ing could also be involved [4].

Here we present preliminary results showing that the morphological features (circular depression and topographical ring) beyond the crater rim of the Bosumtwi crater are the remnants of an ejecta rampart and thus, that the Bosumtwi crater is a rampart crater.

Methods: The Bosumtwi crater in Ghana was ana-lyzed in detail on the base of remote sensing data. A GIS environment was implemented for geospatial analyses of digital elevation models (DEMs) and mul-tispectral analyses. Digital elevation data from the Shuttle Radar Topography Mission (SRTM) with a

resolution of 1 arc-second (~30 m at the equator) were used as base data to derive catchment areas and drain-age networks in the surrounding of Bosumtwi crater (Fig. 1a). For morphometric comparisons, we deter-mined the mean elevation of the topography at defined radial distances from crater center (bins of 200 m) in order to construct an averaged radial elevation profile (Fig. 1b). In the process, the southern margin was cropped and not incorporated into the analysis because this area is strongly affected by the northern flank of the NE–SW trending Obuom Mountain Range (Fig 1a).

Fig.1: SRTM elevation data and derived averaged radial elevation profile of Bosumtwi crater (A/B) show a circular depression followed by an elevated ring beyond the crater rim of Bosumtwi crater; CTX DTM and derived averaged radial elevation profile of the Martian DLE crater Steinheim (C/D) [data from 11] show similar results including moat and rampart.

In addition, we used the principal component anal-

ysis (PCA) on the base of Landsat and ASTER data to emphasize variations in the multispectral data and thus bring out strong patterns in the dataset (Fig. 2b). In further steps, we compared the results with a Martian rampart crater of similar size. The Martian crater Steinheim (190.65°E 54.57°N)[11] is a young 11.2 km diameter double-layered ejecta (DLE) crater in Arca-dia Planitia and has an excellent coverage of high-resolution image data. We used the Ames Stereo Pipe-line to build a high-resolution digital elevation model (DEM) based upon CTX imagery and MOLA (Mars

2583.pdfLunar and Planetary Science XLVIII (2017)

Orbiter Laser Altimeter) data for further detailed anal-yses of the ejecta morphology of Steinheim crater [10], including a radial elevation profile and the determina-tion of a hypothetic drainage pattern and catchment areas (Fig. 1c, d).

Fig.2: A) Mapped breccias in the surrounding of Bosumtwi crater (data from [2]: red = suevites (polymict, glass-bearing breccias); white circles = polymict clastic impact breccias; white triangles = polymict clastic breccias (genetically un-classified, likely impact-related); black circles = polymict and monomict clastic breccias (unspecified, possibly impact-related). B) False colour RGB image of the Principal com-ponent analysis (PCA) on the base of ASTER VNIR data.

Fig.3: Derived drainage pattern and accordant catchment areas for Bosumtwi crater (A) and the Martian Steinheim crater (B).

Results and Discussion: The SRTM elevation da-ta and the derived averaged radial elevation profile show a circular depression followed by an elevated ring beyond the crater rim of Bosumtwi crater (Fig. 1a, b). The morphological characteristics of this topo-graphic pattern show striking similarities with regard to position, dimension, and shape to those of Martian DLE craters, which typically possess depressions (called moats) and subsequent broad, elevated ridges (called ramparts) [11] (Fig. 1c, d). In the case of Mar-tian craters, this morphological trend is due to the dis-tribution of the ejecta blanket. Indeed, Bosumtwi also shows a lot of evidence for ejecta deposits within this area including a multitude of impact-related and possi-bly impact-related breccias (Fig. 2a). Although the

multispectral data of this area provide no clear signa-ture for possible ejecta deposits due to the dense rain-forest cover, a weak halo-like signal surrounds the impact crater possibly indicating a similar composition of the moat and rampart area (Fig. 2b). Following this approach, it is conspicuous that the drainage network around Bosumtwi crater is showing a circular pattern in the slightly depressed annular zone (Fig. 3a). As-suming Bosumtwi crater as the result of a lunar like impact event, the thickness of the ejecta blanket would decrease with increasing radial distance to crater center as the result of ballistic sedimentation and erosion. Under such circumstances, a radial drainage pattern would be expected in contrast to the actual concentric pattern. As a thought experiment, we have generated a hypothetical drainage network for the Martian DLE crater Steinheim. Interestingly, the drainage pattern of Steinheim crater shows a lot of similarities to Bosumtwi including a local watershed along the crest-line of the rampart and a concentric discharge pattern in the moat area (Fig. 3b).

Conclusions and Outlook: The morphological features (circular depression and topographical ring) beyond the crater rim of Bosumtwi crater as well as the ejecta distribution and drainage pattern show many similarities to Martian rampart craters. Therefore, we suggest that Bosumtwi crater shows the eroded rem-nants of an ejecta rampart, possibly similar to the Ries crater in Germany [12], and thus, is a terrestrial ram-part crater. In future works, this assumption will be proven in more detail by using TanDEM-X imagery of higher resolution and further multispectral data anal-yses.

References: [1] Scholz, A.C. et al. (2002) Geology, vol. 30, pp. 939–942. [2] Koeberl, C. and Reimold, W.U. (2005) Jahrbuch der Geologischen Bundesanstalt, vol. 145, pp. 31–70. [3] Koeberl, C. et al. (1997) Geochim. Cosmochim. Acta, vol. 61, pp. 1745–1772. [4] Reimold, W.U. and Koeberl, C. (2014) Journal of African Earth Sci-ences, vol. 93, pp. 57-175. [5] Turner, B.F. et al. (1996) Limnol. Oceanogr., vol. 41, pp. 1415–1424. [6] Koeberl, C. et al. (2007) Meteoritics & Planetary Science, vol. 42, pp. 477–896. [7] Jones, W.B. et al. (1981) Ghana. Geol. Soc. Am. Bull., vol. 92, pp. 342–349. [8] Wagner, R. et al. (2002) In Meteorite Impacts in Precambrian Shields. Impact Stu-dies, vol. 2: Plado, J., Pesonen, L.J. (Eds.), Springer, Berlin, Heidelberg, pp. 189–210. [9] Pesonen, L.J. et al (2003) Yearbook of the Austrian Geological Survey, Vienna, vol. 143, pp. 581–604. [10] Moratto Z.M. et al. (2010) 41st Lunar and Planetary Science Conference, abstract #2364. [11] Wulf G. and Kenkmann T. (2015). Meteoritics & Planetary Science, vol. 50, pp. 173–203. [12] Sturm S. et al. (2013) Geology, vol. 41(5), pp. 531-534.

2583.pdfLunar and Planetary Science XLVIII (2017)