Knowledge-Driven GeoAI: Integrating Spatial Knowledge into Multi-Scale Deep Learning for Mars Crater Detection

This paper introduces a new GeoAI solution to support automated mapping of global craters on the Mars surface. Traditional crater detection algorithms suffer from the limitation of working only in a semiautomated or multi-stage manner, and most were developed to handle a specific dataset in a small subarea of Mars’ surface, hindering their transferability for global crater detection. As an alternative, we propose a GeoAI solution based on deep learning to tackle this problem effectively. Three innovative features are integrated into our object detection pipeline: (1) a feature pyramid network is leveraged to generate feature maps with rich semantics across multiple object scales; (2) prior geospatial knowledge based on the Hough transform is integrated to enable more accurate localization of potential craters; and (3) a scale-aware classifier is adopted to increase the prediction accuracy of both large and small crater instances. The results show that the proposed strategies bring a significant increase in crater detection performance than the popular Faster R-CNN model. The integration of geospatial domain knowledge into the data-driven analytics moves GeoAI research up to the next level to enable knowledge-driven GeoAI. This research can be applied to a wide variety of object detection and image analysis tasks.

ExoMars Raman Laser Spectrometer: A Tool for the Potential Recognition of Wet-Target Craters on Mars

In the present work, near-infrared, laser-induced breakdown spectroscopy, Raman, and X-ray diffractometer techniques have been complementarily used to carry out a comprehensive characterization of a terrestrial analogue selected from the Chesapeake Bay impact structure (CBIS). The obtained data clearly highlight the key role of Raman spectroscopy in the detection of minor and trace compounds, through which inferences about geological processes occurred in the CBIS can be extrapolated. Beside the use of commercial systems, further Raman analyses were performed by the Raman laser spectrometer (RLS) ExoMars Simulator. This instrument represents the most reliable tool to effectively predict the scientific capabilities of the ExoMars/Raman system that will be deployed on Mars in 2021. By emulating the analytical procedures and operational restrictions established by the ExoMars mission rover design, it was proved that the RLS ExoMars Simulator can detect the amorphization of quartz, which constitutes an analytical clue of the impact origin of craters. Beside amorphized minerals, the detection of barite and siderite, compounds crystallizing under hydrothermal conditions, helps indirectly to confirm the presence of water in impact targets. Furthermore, the RLS ExoMars Simulator capability of performing smart molecular mappings was successfully evaluated.

Methane Seepage on Mars: Where to Look and Why

Methane on Mars is a topic of special interest because of its potential association with microbial life. The variable detections of methane by the Curiosity rover, orbiters, and terrestrial telescopes, coupled with methane's short lifetime in the martian atmosphere, may imply an active gas source in the planet's subsurface, with migration and surface emission processes similar to those known on Earth as "gas seepage." Here, we review the variety of subsurface processes that could result in methane seepage on Mars. Such methane could originate from abiotic chemical reactions, thermogenic alteration of abiotic or biotic organic matter, and ancient or extant microbial metabolism. These processes can occur over a wide range of temperatures, in both sedimentary and igneous rocks, and together they enhance the possibility that significant amounts of methane could have formed on early Mars. Methane seepage to the surface would occur preferentially along faults and fractures, through focused macro-seeps and/or diffuse microseepage exhalations. Our work highlights the types of features on Mars that could be associated with methane release, including mud-volcano-like mounds in Acidalia or Utopia; proposed ancient springs in Gusev Crater, Arabia Terra, and Valles Marineris; and rims of large impact craters. These could have been locations of past macro-seeps and may still emit methane today. Microseepage could occur through faults along the dichotomy or fractures such as those at Nili Fossae, Cerberus Fossae, the Argyre impact, and those produced in serpentinized rocks. Martian microseepage would be extremely difficult to detect remotely yet could constitute a significant gas source. We emphasize that the most definitive detection of methane seepage from different release candidates would be best provided by measurements performed in the ground or at the ground-atmosphere interface by landers or rovers and that the technology for such detection is currently available. Key Words: Mars-Methane-Seepage-Clathrate-Fischer-Tropsch-Serpentinization. Astrobiology 17, 1233-1264.

Giant polygons and mounds in the lowlands of Mars: signatures of an ancient ocean?

This paper presents the hypothesis that the well-known giant polygons and bright mounds of the martian lowlands may be related to a common process-a process of fluid expulsion that results from burial of fine-grained sediments beneath a body of water. Specifically, we hypothesize that giant polygons and mounds in Chryse and Acidalia Planitiae are analogous to kilometer-scale polygons and mud volcanoes in terrestrial, marine basins and that the co-occurrence of masses of these features in Chryse and Acidalia may be the signature of sedimentary processes in an ancient martian ocean. We base this hypothesis on recent data from both Earth and Mars. On Earth, 3-D seismic data illustrate kilometer-scale polygons that may be analogous to the giant polygons on Mars. The terrestrial polygons form in fine-grained sediments that have been deposited and buried in passive-margin, marine settings. These polygons are thought to result from compaction/dewatering, and they are commonly associated with fluid expulsion features, such as mud volcanoes. On Mars, in Chryse and Acidalia Planitiae, orbital data demonstrate that giant polygons and mounds have overlapping spatial distributions. There, each set of features occurs within a geological setting that is seemingly analogous to that of the terrestrial, kilometer-scale polygons (broad basin of deposition, predicted fine-grained sediments, and lack of significant horizontal stress). Regionally, the martian polygons and mounds both show a correlation to elevation, as if their formation were related to past water levels. Although these observations are based on older data with incomplete coverage, a similar correlation to elevation has been established in one local area studied in detail with newer higher-resolution data. Further mapping with the latest data sets should more clearly elucidate the relationship(s) of the polygons and mounds to elevation over the entire Chryse-Acidalia region and thereby provide more insight into this hypothesis.