Scientists have developed a new machine-learning tool that could be of help in the search for signs of life on Mars and other celestial bodies in space.

While it’s hard and expensive to even collect samples from other planets, scientists rely on remote sensing methods in order to continue their search for alien life. But if they could find a better way to refine their investigation, the result could be huge.

A team of scientists led by Kim Warren-Rhodes of the Search for Extraterrestrial Intelligence (SETI) institute in California has mapped the sparse lifeforms that dwell in salt domes, rocks, ad crystals in the Salar de Pajonales, a salt flat on the boundary of the Chilean Atacama Desert and Altiplano, or high plateau.

Waren Rhodes’ team as well as that o Michael Phillips from the John Hopkins University Applied Physics Laboratory and University of Oxford researcher, Freddie Kalaitzis in order to train machine learning models so as to recognize the patterns and rules associated with the distribution of life across the aforementioned regions.

Using this training, the models can spot the same patterns and rules for a wide range of landscapes including those that may be encountered on other rocky planets in the solar system.

The team discovered that their system could make use of AI technology to locate and detect biosignatures up to 87.5% o the time compared t the 10% success rate achieved by random searches.

Implementing such technology could also help decrease the area needed for a search by over 97%, which would help scientists to focus their limited resources in their hunt for a potential chemical trace of life or biosignatures.

“Our framework allows us to combine the power of statistical ecology with machine learning to discover and predict the patterns and rules by which nature survives and distributes itself in the harshest landscapes on Earth,” Warren-Rhodes said in a statement. “We hope other astrobiology teams adapt our approach to mapping other habitable environments and biosignatures.”

The researchers also made it known that such learning tools could be used for robotic planetary missions such as NASA’s Perseverance rover which is currently working on Mars’ Jezero Crater, in search of alien life.

“With these models, we can design tailor-made roadmaps and algorithms to guide rovers to places with the highest probability of harboring past or present life  —  no matter how hidden or rare,” Warren-Rhodes explained. 

Using Earth as a mock-up for Mars

The reason why the scientists chose Salar De Pajonales as their testing ground for the machine learning model was because it has a similarly harsh environment and landscape as Mars.

The region is a high-altitude dry salt lakebed that is well-exposed to a high degree of ultraviolet radiation.

Despite being considered highly inhospitable to life, however, Salar de Pajonales still harbors some living things. 

The team took over 8,000 images and more than 1,000 samples from the location in order to detect photosynthetic microbe that lives within the region’s salt domes, rocks, and alabaster crystals.

Pigments secreted by these microbes represent a possible micro signature on NASA’s “ladder of life detection,” which the agency designed to guide the search for life outside of Earth within the practical constraints of robotic space missions.  

In their examination, the team further examined the Salar De Pajonales using drone imagery which is analogous to images of Martian terrain captured by the High-Resolution Imgaing Experiment (HIRISE) camerathat was on the Mars Reconnaisance Orbiter.

Data from this allowed the research team to determine that microbial life on the test site is not randomly distributed but concentrated in biological hotspots which are linked to the availability of water.

The team also trained convolutinal neual networks (CNNs) to recognize ad predict large geologic features at Salar de Pajonales with some of the features including patterned ground annd polygonal networks which are also found on Mars’ surface.

The CNN was also trained to spot and predict smaller microhabitats most likely to contain biosignatures.

For now, the team will continue with their trainings at the site while aiming to test the CNN’s ability to predict the location and distribution of ancient stromatolite fssiles and salt-tolerant microbiomes.

This should help it to learn if the rules it uses in this search could also apply to the hunt for biosignatures in other similar natural systems. 

Aterward, the team will then map hot springs, froen permafrost-coered soils, and the rocks in dry valleys in an attempt to teach the AI to hone in on potential habitats in other extreme environments on our planet before exploring those on other planets

The team’s research was published this month in the journal Nature Astronomy.