Verified by 
Using AI to Search for Life on Mars and Other Planets
SETI and NASA use AI machine learning to search for biosignatures beyond Earth.
Posted March 9, 2023 Reviewed by Lybi Ma
Can the pattern recognition capabilities of artificial intelligence (AI) help in the search for biosignatures in the universe? An interdisciplinary team of over 50 scientists from more than 17 institutions led by astrobiologists on the SETI Institute NASA Astrobiology Institute (NAI) team published new peer-reviewed research in Nature Astronomy showcasing an AI machine learning algorithm to search for life on Mars and beyond.
“In the search for biosignatures on Mars, there is an abundance of data from orbiters and rovers to characterize global and regional habitability, but much less information is available at the scales and resolutions of microbial habitats and biosignatures,” wrote the lead author Kim Warren-Rhodes, Senior Research Scientist at the SETI Institute, along with Michael Phillips, a postdoctoral researcher at The Johns Hopkins University Applied Physics Laboratory, Freddie Kalaitzis, a Senior Research Fellow at the University of Oxford, and other researchers.
The Goldilocks Zone, Just Right for Life?
The Goldilocks zone or habitable zone, where conditions are not too hot and not too cold to support life, is just right. A habitable zone is the planet’s distance from a star that enables liquid water on the surface due to its atmosphere having just-the-right conditions. The Earth is in the habitable zone in our solar system that consists of the Sun and the planets that orbit around it. Are there other planets in the habitable zone in our solar system?
In a different research study published in 2016 by Thomas Clune, Ph.D., and others at NASA Goddard Institute for Space Studies, the scientists posit that Venus may have once had a climate that supported liquid water on its surface for roughly two billion years, and that there are Venus-like planets outside of the Earth’s solar system, or exoplanets, that are prime candidates for searching for life outside of our planet.
“Present-day Venus is an inhospitable place with surface temperatures approaching 750 K and an atmosphere 90 times as thick as Earth's,” wrote Clune and fellow research colleagues. “Billions of years ago the picture may have been very different.”
Flowing Water on Present-Day Mars
Observations from NASA's Mars Reconnaissance Orbiter reveal the seasonal formation of dark streaks that repeatedly appears on select craters on Mars during the warmer months of late spring through summer, called the recurring slope lineae (RSL). Saltwater flows on Mars were hypothesized in 2011 from these observations. Four years later in 2015, NASA confirmed evidence that liquid water flows intermittently on present-day Mars.
“Our findings strongly support the hypothesis that recurring slope lineae form as a result of contemporary water activity on Mars,” wrote lead author Lujendra Ojha at the Georgia Institute of Technology (Georgia Tech) in Atlanta, along with research colleagues who co-authored the 2015 study published in Nature Geoscience.
"Our quest on Mars has been to 'follow the water,' in our search for life in the universe, and now we have convincing science that validates what we've long suspected," said John Grunsfeld, Ph.D., Associate Administrator for the Science Mission Directorate at NASA Headquarters in Washington, D.C. and five-time astronaut (on board the Endeavour in 1995, Atlantis in 1997 and 2009, Discovery in 1999, and Columbia 2002 space shuttle missions) in a NASA 2015 report. "This is a significant development, as it appears to confirm that water, albeit briny, is flowing today on the surface of Mars.”
If There’s Water, Is There Life too?
The tantalizing evidence of contemporary water flows on Mars begs the question: Is there life too?
“In the search for biosignatures on Mars, there is an abundance of data from orbiters and rovers to characterize global and regional habitability, but much less information is available at the scales and resolutions of microbial habitats and biosignatures,” wrote the SETI Institute NAI team of astrobiologists and their research colleagues. “Understanding whether the distribution of terrestrial biosignatures is characterized by recognizable and predictable patterns could yield signposts to optimize search efforts for life on other terrestrial planets.”
Using AI to Find the Biosignature Needle in the Haystack
The researchers decided to utilize the predictive powers of AI machine learning to search for signs of life. Scientists conducting experiments in-person on Mars is not a current option; the team needed to find a similarly harsh environment that is habitable, but hostile to many living organisms. They decided to use samples from Salar de Pajonales, a dry, calcium-sulfate salt lakebed located in the Chilean High Andes at over 3,500 meters altitude, as a surrogate for Mars.
The team of scientists collected samples from over 7,700 images and 1,100 samples from the arid lakebed. They conducted experiments to search for photosynthetic microbes that contain pigments that may prove to be a candidate biosignature.
The AI algorithm consists of convolutional neural networks (CNNs), that were trained using the data collected to predict large-scale geologic features and micro-habitats that have the highest likelihood of having biosignatures. The scientists tested their Al machine learning’s prediction ability.
“Artificial intelligence-machine learning models successfully identified geologic features with high probabilities for containing biosignatures at spatial scales relevant to rover-based astrobiology exploration,” the scientists wrote.
Copyright © 2023 Cami Rosso All rights reserved.
