A Fluvio-Lacustrine Environment Preserved in the Jezero Crater Inlet Channel, Neretva Vallis, Mars

Martian fluvial valleys provide evidence for the surface flow of liquid water, making them a key target for rover‐based investigations of ancient habitability. The Mars 2020 Perseverance rover spent ∼85 sols exploring the Bright Angel formation, exposed across the floor of Neretva Vallis: the western inlet channel of Jezero crater. This study documents the sedimentology and stratigraphy of the Bright Angel formation to reconstruct its depositional setting. The unit preserves a concave‐up bedding structure consistent with a young channel‐fill deposit, rather than an older unit exposed by incision of Neretva Vallis. The lower stratigraphy displays a fining‐up sequence from coarse‐grained sediments up to pebble‐conglomerates (the Tuff Cliff member) into a ≥10‐m‐thick succession of laminated mudstone (the Walhalla Glades member), interpreted as a transgressive sequence recording the onset of lacustrine conditions in Neretva Vallis. Lenses of matrix‐supported granule‐conglomerate adjacent to the valley wall (the Fern Glen Rapids member) may preserve locally derived debris flows entering the lake. These are overlain by a polymict, matrix‐supported, boulder‐conglomerate (the Mount Spoonhead member), interpreted as a high‐energy debrite derived from the watershed. The sequence is capped by cross‐stratified sediments (the Serpentine Rapids member), preserving lake margin deposits. The Bright Angel lacustrine sequence occurs ∼10–50 m higher in elevation than the lake level anticipated for the Jezero western delta, requiring an additional period of lacustrine activity. The structure and spatial distribution of the unit leads us to propose that a late‐stage blockage of Neretva Vallis may have facilitated the formation of a perched, valley‐confined lake upstream.

The authors wish to thank the Mars 2020 Perseverance rover science and engineering teams for their hard work in tactical strategic and science operations, without whom this study would not be possible. We also thank the Mastcam-Z, SuperCam, SHERLOC and RIMFAX operations teams, and the Mars 2020 project management for their ongoing efforts in making the mission a success and in furthering our understanding of Mars science. We are very grateful to Martin Gibling and a second anonymous reviewer for their thorough and helpful reviews of this work. A. J. Jones was funded by the Imperial College London President's PhD Scholarship. UK Space Agency Grants ST/Y000153/1 and ST/X002373/1 funded S. Gupta and R. Barnes. G. Paar and C. Traxler were funded by FFG ASAP Mars 4-D Grants 911911 and FO999911911. J. Hurowitz was funded by NASA Jet Propulsion Laboratory subcontract 1529702. B. Wogsland was funded by the SHERLOC WATSON instrument team. A. Broz, B. Horgan, B. Garczynski, A. Klidaras and J. Bell were funded by a Jet Propulsion Laboratory subcontract 1511125 to Arizona State University. This work was carried out by K. Stack, J. Maki and K. Farley at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004). J. I. Simon was funded by the NASA Science Mission Directorate and the Mars Exploration Program. SAC acknowledges NASA Grant NNH13ZDA018O for supporting this work. A. Murphy was funded by the Mars 2020 Program through a Planetary Science Institute subcontract (1641753). O. Gasnault was funded by CNRS and CNES. The Australian Research Council funded D. Flannery.