Identification of a New Outflow in the Syrtis Major Region, Mars

Introduction: Outflow channels on Mars are formed by very energetic floods suggesting catastrophic episodes of concentrated water flows [1]. These floods comes from a gigantic release of water such as in terrestrial glacial surge, but their exact origin is debated. Outflow channels of the Xanthe Terra region east of the Tharsis area are all of the Late Hesperian epoch and their source regions are formed by chaotic terrains of unknown origin. Most models suggest a formation by groundwater release either by aquifers, permafrost melting from increasing geothermal flux or magmatic activity, or eventually impact crater effect [1,2,3]. Nevertheless, outflows such as Mangala valles or the recently discovered Athabasca Valles have similar channels characteristics but a source area consisting of a single fracture likely due to tectonicvolcanic activity [4,5]. The identification of new area of similar floods is important to complete the different types of outflow formation and understand better the relationships with magmatic activity. Observational evidence for floods: HRSC images of NE Syrtis Major provides high resolution (12 to 22 m/pixel) images of the lava plains. Close up shows surface features different than usual lava flows. The first feature consists of elongated striations (Fig. 2). The striations, or grooves, are oriented in the NS direction and present narrow width compared to their length of several kilometers. These grooves typical are of erosion into resistant bedrock. They may be created by glacier abrasion, wind erosion or water flows. Wind erosion forms features such as yardangs and is unlikely to explain grooves here. Indeed, many grooves are curved while yardangs are usually very straight. Glaciers abrasion forms grooves at different scales with other features associated such as drumlins and moraines. In contrast, a water flood could create grooves similar to those observed in outflow channels such as Kasei Valles. Such flood would not affect the Noachian hill if the flow is thin and slow (given the slope) and surrounds the hill. The second feature observed is a tear-dropped island (Fig. 2). This feature is very similar in shape to the islands observed in classical outflow channels such as Ares Valles [1]. These features are due to the erosion by energetic water flows stopped by an obstacle such as an impact crater. This forms a zone of non erosion, or at least of less erosion than in the middle of the flood, in the “shadow” behind the crater. This feature is thus a main argument in favor of liquid water flows over the smooth plains. The orientation of the island also gives a North to South direction of flow, which is consistent with the orientation of erosional grooves. Southeast of the erosional grooves, we can observe the presence of a deep valley about 100 km long. At low resolution, this valley resembles to a lava tube as often seen over Martian lava flows. Nevertheless, close-up over this valley shows clear characteristics of fluvial valleys. The main valley is connected to a few tributaries and are also frequently sinuous. Both characteristics are consistent with water flows better than with lava tubes. Thus, the valleys might be due to a higher incision of the flow at this location while no incision occurred in the western part. This hypothesis is consistent with local slopes. At the beginning of the valley, there is a gentle step in the topography corresponding to the change of the slope from a NorthSouth direction. The slope is here of 0.8° where the valley occurs, thus more than in the figure 2 area where it is lower than 0.1°. This may explain that the flood was not canalized in the flattest part of the plain, and that it is canalized after this step. East of the figure, the valley disappears progressively. Here the topography becomes more flat again, suggesting a widening of the flow without any valley at that place. Unfortunately, we miss data at the eastern end of the flood, thus making impossible the detection of possible terminal deposits.