A recent study suggests that artificial intelligence (AI) could help scientists locate the ideal location to look for life on Mars and other frigid worlds.
A team of astrobiologists created an AI model and evaluated its ability to detect scarce life hidden away in salt domes, boulders, and crystals at Salar de Pajonales on the border of Chile’s Atacama Desert and Altiplano – one of the planet’s driest regions, resembling the features of a Martian surface.
Pajonales is a high altitude (3,541 m) hyperarid, dry salt lakebed that is unsuitable to many living forms yet nonetheless inhabited.
The findings, published in the journal Nature Astronomy, demonstrated that the AI model assisted scientists in locating and detecting biosignatures (any feature that offers evidence of past or current life) up to 87.5 percent of the time.
It also decreased the region the researchers needed to search by up to 97%, indicating the AI model’s ability to find signs of life on distant planets one day.
When searching for life beyond Earth, researchers now have few options to collect samples on Mars or elsewhere, or to use remote sensing tools. The new AI model will assist scientists in determining the precise location to search for life on distant worlds.
“We hope other astrobiology teams adapt our approach to mapping other habitable environments and biosignatures,” said lead researcher Kim Warren-Rhodes, Senior Research Scientist at SETI Institute.
“With these models, we can design tailor-made roadmaps and algorithms to guide rovers to places with the highest probability of harbouring past or present life — no matter how hidden or rare,” Warren-Rhodes added
The team, which included members of NASA’s Astrobiology Institute (NAI), gathered over 7,765 photographs, 1,154 samples, and tested tools to find photosynthetic bacteria living within salt domes, rocks, and alabaster crystals.
The study’s findings demonstrate (statistically) that microbial life at the Pajonales terrestrial analogue site is concentrated in patchy biological hotspots that are closely linked to water availability at km to cm scales.
Following that, the team trained CNNs to recognise and forecast macro-scale geologic features at Pajonales – some of which, such patterned ground or polygonal networks, are also found on Mars – and micro-scale substrates (or’micro-habitats’) most likely to contain biosignatures.
The researchers studied ways to combine a UAV/drone with ground-based rovers, drills, and instruments (e.g., VISIR on ‘MastCam-Z’ and Raman on ‘SuperCam’ on the Mars 2020 Perseverance rover).
“While the high-rate of biosignature detection is a central result of this study, no less important is that it successfully integrated datasets at vastly different resolutions from orbit to the ground, and finally tied regional orbital data with microbial habitats,” said Nathalie A. Cabrol from the SETI Institute NAI team.