Analyses of dendritic ridges within Antoniadi crater, Mars, from CaSSIS and HiRISE data

<p><strong>Introduction</strong></p> <p>Antoniadi basin is a 330 km diameter Noachian basin localized in the eastern Arabia Terra that contains a network of dendritic ridges. Branched, dendritic ridges, such as these can form by a variety of processes including the inversion of fluvial deposits, thus potentially highlighting aqueous processes of interest for understanding Mars&#8217; climate evolution. Here, we test this hypothesis by analyzing in details data from Colour and Stereo Surface Imaging System (CaSSIS), High Resolution Imaging Science Experiment (HiRISE) and High Resolution Stereo Camera (HRSC).</p> <p><strong>Age of landforms</strong></p> <p>Antoniadi&#8217;s interior plains are filled by deposits interpreted to be Hesperian or younger and volcanic in origin, possibly coeval with the volcanic episodes of the nearby Syrtis Major Planum [1]. The dendritic ridges lie on these volcanic plains to the south of a 28 km-diameter well-preserved crater (Fig. 1). Although this crater predates the branched ridges it is devoid of any fluvial erosion and appears mostly similar to many fresh, Amazonian age craters. While some erosion has affected the ridges, there is no indication of exhumation of these ridges from putative layers above the current plains surface. The crater size frequency distribution of the volcanic plains gives a model age of 2.5&#177;0.5 Gy, using diameters of 200 m to 2 km. Thus, the branched ridges are Early Amazonian or younger.</p> <p><img src="" alt="" width="600" height="594" /></p> <p><strong>Figure. 1: </strong>Four Cassis images projected on CTX mosaic. The fresh crater to the north has been covered by fresh lava flows which themselves have been covered by the dendritic features.</p> <p><strong>Morphometry and topography</strong></p> <p>Individual ridges are easily recognizable with their dark tone in visible images compared to surrounding plains (Fig. 2), which display a slightly lighter tone. The ridges are organized as a dendritic network reaching a Strahler order of 4, i.e. the degree of hierarchy from the primary branches counting the number of junctions. Assuming tributary flows, this organization indicates a northward flow direction. However, the local slope is of 0.2&#176; toward South, and thus contrary to the apparent network organization (assuming tributary flows). The branched ridges are also present on some of the lowest areas of these plains, but there is no evidence of terminal fans nor of erosional features affecting the plains in their more elevated areas.</p> <p>The texture of ridges at HiRISE scale (25cm/pixel) is rubbly with the occurrence of blocks up to ~1 m in size and a complete lack of layering in all HiRISE images where they are present unlike sedimentary strata usually observed within inverted channels on Mars. A HiRISE elevation model shows that branched ridges are between 1-5 m high, although erosion can explain some variations in height among the ridges. Ridges are up to 10 km long and 10-200 m wide without any obvious trends in width, i.e. ridges are not wider with increasing Strahler order. The order 1 branches are also peculiar in plan-view shape. They frequently display lobate shapes and are up to 100 m wide, which is wider than the ridge measured after the junction with several order 1 branches. North of the main pattern, a 500 m wide sinuous ridge present in the middle of the plain displays the same internal texture (rubbly) suggesting that it formed by the same process (white arrow at the top of Fig. 1).</p> <p><strong>&#160;</strong><img src="" alt="" /></p> <p><strong> </strong></p> <p><strong>Figure 2:</strong> Close-up on the dendritic network with CaSSIS images.</p> <p><strong>Conclusion</strong></p> <p>Previous assessments of these landforms favored an origin as inverted channels [2]. Yet, our observations show many inconsistencies with this interpretation: (i) The rubbly texture lacks any layering at meter scale, a typical feature of inverted channels observed elsewhere on Mars [e.g., 3, 4]. (ii) The order 1 branches display a lobate shape and a larger width than higher order branches unlike river flows as observed on Earth. (iii) There is no increase in width from degree 1 branches of the network towards the north as would be expected for channels with increasing discharge rates downstream. (iv) The slope towards the south is contrary to the inferred flow direction to the north assuming a tributary network. Wrinkle ridges may be evidence for post-depositional changes in topography, but these uplifts appear localized (Fig. 1). Thus, the detailed analysis of these branched ridges shows characteristics difficult to reconcile with inverted fluvial channels. Subglacial drainages are known to locally flow against topography, but rarely display a dendritic pattern [5].<strong> </strong>Assuming that deposition occurred along the current slope from north to south, the organization of the network requires a control by distributary channels rather than tributary channels. Distributary fluvial channels are possible for fluvial flows, but generally limited to braiding regimes or deltaic deposits, of which no further evidence is observed here. The lobate digitate shapes of the degree 1 branches are actually more in line with deposits of viscous flows, thus as terminal branches (Fig. 3). Such an interpretation is consistent with lava flows or mudflows that formed along the current topography. This conclusion would explain the presence of these landforms in an Early Amazonian plain lacking any evidence of fluvial activity, including the fresh crater that is stratigraphically older than the landforms studied.</p> <p><strong>Acknowledgments:</strong> French authors are supported by the CNES. The authors wish to thank the spacecraft and instrument engineering teams. CaSSIS is a project of the University of Bern and funded through the Swiss Space Office via ESA&#8217;s PRODEX. The instrument hardware development was also supported by the Italian Space Agency (ASI) (agreement no. I/018/12/0), INAF/Astronomical Observatory of Padova, and the Space Research Center (CBK) in Warsaw. Support from SGF (Budapest), the Univ. of Arizona (Lunar and Planet. Lab.) and NASA are gratefully acknowledged.</p> <p><strong>References&#160;:</strong> [1]&#160;Tanaka K. L. et al. (2014) USGS map 3292. [2]&#160;Zaki et al. (2019) EPSC-DPS2019-244-1. [3]&#160;Burr et al.&#160; (2010) <em>JGR-Planets, </em>115, E07011. [4]&#160;Mangold et al. (2008) <em>JGR-Planets, </em>113, E08009. [5]&#160;Menzies, J (2002), Modern and Past Glacial Environments, Oxford.&#160;</p>