The banded terrain on northwestern Hellas Planitia: New observations and insights into its possible formation

Northwestern Hellas Planitia hosts landforms that are unique on Mars, e.g., the so called honeycomb and banded (aka “taffy pull”) terrains. Recently, robust formation models for the ∼6 km large honeycomb depressions involving salt or ice diapirism have been formulated. However, the nature of the banded terrain, a ∼30,000 km² area characterized by a decameter- to kilometer-scale pattern of curvilinear troughs, has remained elusive. While previous interpretations range from deep-seated, honeycomb-related outcrops to a younger veneer, recent reports of putative periglacial features (e.g., potential thermokarst) strongly indicate it to be a relatively thin, volatile-related surface unit. In order to further constrain the origin and nature of the banded terrain, we investigated the northwestern Hellas basin floor employing various datasets. We mapped the banded terrain's extent at high precision, showing that it partially superposes the honeycomb terrain, but also occurs up to ∼240 km away from it. Via stratigraphic analyses and crater size-frequency measurements, we bracketed the age of the banded terrain between ∼1.9 and ∼3.7 Ga. Furthermore, the banded terrain can be differentiated into two types, ridged and creviced, with the former predominantly occurring among the lowest reaches of the terrain's ∼2 km topographic extent. We also produced a grid map (2 × 2 km box size) of the entire banded terrain and identified no large-scale (> 25 km) band pattern and no correlation between local slope and band orientation. Because of this, we submit that regional tectonics or gravity-driven flow down modern topography are unlikely to have played decisive roles for banded terrain formation. Instead, we observed numerous locations, where band slabs appear to have broken off and subsequently rotated, as well as “cusps” that seem to have resulted from buckling. Based on this, we suggest that the banded terrain experienced both, ductile deformation as well as brittle failure on or near the surface. Despite certain similarities, neither salt (as salt glaciers), lava sheets, or land-based glaciers are in agreement with the extensive curvilinear texture and topographic/geologic setting of the banded terrain. Ice shelf margins, on the other hand, can produce surface textures akin to the banded terrain in both form and scale, even including cusps and broken off, rotated blocks. However, an ice-covered sea between 1.9 and 3.7 Ga ago is not indicated by the geologic inventory of the Hellas basin, which previous investigations found to lack any landforms indicative of a standing body of water. Instead, we identified several sinuous ridges terminating at plains covered by smaller, braiding ridges, which we interpret as eskers and glacial sandurs, respectively. As both are embayed and partially covered by the banded terrain, we tentatively propose an alternative, subglacial model of the banded terrain having formed as wet till that was viscously deformed according to the stress fields created by the ice overburden pressure in conjunction with bed topography. Although this formation model remains inconclusive, it is in agreement with climate models suggesting obliquity excursions and a denser, early Amazonian atmosphere to have caused ice accumulation in the adjacent northwestern Hellas basin rim, thus potentially enabling flow onto the floor entailing subglacial banded terrain formation.