Combination of MRO SHARAD and deep-learning-based DTM to search for subsurface features in Oxia Planum, Mars

¹ Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ), Kerun B Bld, Yutang Street, Guangming District, Shenzhen 518107, PR China
e-mail: xiongsiting@gml.ac.cn
² Mullard Space Science Laboratory, Department of Space and Climate Physics, University College London, Holmbury St Mary, Dorking, Surrey, RH5 6NT, UK
³ Planetary Sciences and Remote Sensing Group, Department of Earth Sciences, Freie Universität Berlin, Malteserstr. 74-100, Berlin 12249, Germany
⁴ Institute for Advanced Study, Shenzhen University, No. 3688, Nanhai Avenue, Nanshan District, Shenzhen 518060, PR China
⁵ Space and Earth Interdisciplinary Center, Shenzhen University, No. 3688, Nanhai Avenue, Nanshan District, Shenzhen 518060, PR China
⁶ College of Civil and Transportation Engineering, Shenzhen University, No. 3688, Nanhai Avenue, Nanshan District, Shenzhen 518060, PR China

Received: 20 February 2023
Accepted: 2 June 2023

Abstract

Context. Oxia Planum is a mid-latitude region on Mars that attracts a great amount of interest worldwide. An orbiting radar provides an effective way to probe the Martian subsurface and detect buried layers or geomorphological features. The Shallow radar orbital radar system on board the NASA Mars reconnaissance orbiter transmits pulsed signals towards the nadir and receives returned echoes from dielectric boundaries. However, radar clutter can be induced by a higher topography of the off-nadir region than that at the nadir, which is then manifested as subsurface reflectors in the radar image.

Aims. This study combines radar observations, terrain models, and surface images to investigate the subsurface features of the ExoMars landing site in Oxia Planum.

Methods. Possible subsurface features are observed in radargrams. Radar clutter is simulated using the terrain models, and these are then compared to radar observations to exclude clutter and identify possible subsurface return echoes. Finally, the dielectric constant is estimated with measurements in both radargrams and surface imagery.

Results. The resolution and quality of the terrain models greatly influence the clutter simulations. Higher resolution can produce finer cluttergrams, which assists in identifying possible subsurface features. One possible subsurface layering sequence is identified in one radargram.

Conclusions. A combination of radar observations, terrain models, and surface images reveals the dielectric constant of the surface deposit in Oxia Planum to be 4.9–8.8, indicating that the surface-covering material is made up of clay-bearing units in this region.