It’s good that we talk a lot about finding traces of life on Mars, but we also need to know where to look. It’s hard to get to Mars – We want to make sure we make the most of the opportunities available so the trip is not wasted.
But there is a lot of ground to cover. Mars has about the same dry land area as Earth, with one major difference. Throw a rock on the ground, and you’ll likely land somewhere with life. However, the history of life on Mars is a big question mark.
Artificial intelligence and machine learning can make the search for life on Mars less daunting. An international team of researchers led by astrobiologist Kimberly Warren Rhodes of the SETI Institute has shown that these tools can identify patterns hidden in geographic data that could indicate signs of life.
“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 explains.
“We hope that other astrobiology teams will adapt our approach to map other habitable environments and biosignatures. With these models, we can design tailored roadmaps and algorithms to direct rovers to the places most likely to harbor life in the past or present – no matter how hidden they are.” or rare.”
There is one place on Earth that looks strikingly like the barren plains of Mars. This is the Atacama Desert in Chile, one of the driest places on the planet, and it hasn’t seen rain in decades. Even in this inhospitable place, life can be found tucked away in pockets and underground.
Warren Rhodes and her colleagues focused on an area on the border between the Atacama Desert and the Altiplano Plateau called the Salar de Bagonales. This basin is an ancient riverbed and one of the best analogues of the Martian environment on Earth. At 3,541 meters (11,617 ft), it is high in elevation and thus receives high exposure to UV rays. It’s also low in oxygen and extremely dry and salty…but somehow, life can be found there, living in mineral formations.
Over an area of 2.78 square kilometers (1.07 square miles), the researchers carefully took 7,765 images and 1,154 samples, looking for the biosignatures that revealed the presence of photosynthetic microbes. These included the carotenoid pigments and chlorophyll, which color the rocks pink or green.
They also used drones to take aerial photos to simulate images obtained by satellites orbiting Mars and added 3D topographic maps. Then all this information was entered Convolutional Neural Networks (CNNs) to train artificial intelligence to recognize structures in the aquarium that are most likely to be teeming with life.
Interestingly, the CNNs were able to identify the distribution patterns of microbial life in the basin, despite the area’s semi-uniform mineral composition.
The soft metal plaster domes were about 40 percent inhabited, and the ornamental, striped ground with strips of plaster was inhabited by 50 percent. By taking a closer look at the parts of these features that were inhabited, the researchers found the exact habitats. Microbes were strongly attracted to sections of alabaster, a porous form of gypsum that retains water.
The team found that these alabaster microhabitats were “inhabited almost all over the world” and represent the most reliable predictor of biosignatures, indicating that water content is a primary driver of microhabitat distributions.
More relevant to the search for life on Mars, CNNs enabled the researchers to correctly identify biosignatures up to 87.5 percent of the time, compared to up to 10 percent for random searches. This reduced the amount of land they needed for mulching by a whopping 85 to 97 percent.
border frame=”0″allow=”accelerometer; auto start; Clipboard write. gyroscope encoded media; picture in picture; web sharing “allowfullscreen>”.
“For both aerial imagery and ground-based centimeter-scale data, the model demonstrated high predictive power for the presence of geological material that is highly likely to contain biosignatures,” says computer scientist Freddy Kalitzis from the University of Oxford in the United Kingdom.
“The results align well with ground-truth data, as the distribution of biosignatures is closely related to hydrological features.”
Therefore, this approach appears to have multiple benefits. The work has taught us something about life in extreme environments here on Earth and shows promise in identifying life on Mars. And it could help identify other biosignatures here on Earth.
The team plans to try training their CNNs on other biosignatures, such as stromatolites, which are mats of fossilized bacteria that may be billions of years old, and communities of organisms that live in hypersaline environments.
“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 says.
“We hope that other astrobiology teams will adapt our approach to map other habitable environments and biosignatures. With these models, we can design tailored roadmaps and algorithms to direct rovers to the places most likely to harbor life in the past or present – no matter how hidden they are.” or rare.”
Research published in natural astronomy.
“Infuriatingly humble alcohol fanatic. Unapologetic beer practitioner. Analyst.”