To live, life as we know it must exist in a specific habitat. Nonetheless, there are some animals on Earth that appear to be able to thrive under unfavorable conditions, such as those with extremely low temperatures, little water, or oxygen. These kinds of ecosystems are comparable to those on other planets.
There has been a long-running effort to discover life on Mars. Despite having no hard evidence, scientists believed they had discovered life on Mars in 1996.
The use of AI and ML techniques can support the Mars research, according to a recent study led by an international team of more than 50 researchers. With the aid of this technology, hidden patterns in geographic information may be found that point to the possibility of Martian life.
Research History
Mars used to be different from what it is now. Because the sun was faint at initially, Earth and Mars should have been frozen during their early histories. Water, however, was flowing on both worlds, indicating that they must have both had dense atmospheres.
Volcanoes existed on both Earth and Mars, but the Martian ones were never extremely active. Given that volcanic eruptions release significant amounts of water, this knowledge is crucial.
The first spacecraft to report that there is currently no proof of life on Mars was the Viking program in the 1970s. The Mars Surveyor Program was started as part of more extensive study. Between 1996 and 2005, five spacecraft were supposed to be dispatched to Mars. The Mars Global Surveyor, Mars Climate Orbiter, and Polar Lander were among the planned spacecraft.
Polar Lander and the Mars Climate Orbiter were sadly lost. After all the data is gathered, however, researchers still hold out hope that they will be able to offer more details about Mars’ history and its potential to support life.
AI research on bio signatures
An ecological survey of a three km2 area in the Salar de Pajonales basin, which lies at the border of the Chilean Atacama Desert and the South American Altiplano, was conducted as part of the study’s first phase using the AI/ML model under the direction of Dr. Kimberley Warren-Rhodes at the SETI Institute. This examined the microbial dispersion.
On data that it had not been trained on, the final model could locate and identify biosignatures up to 87.5% of the time. This resulted in a 97% reduction in the search area needed to find the successful result. In the future, life on Mars might be found by locating the region’s most likely to have traces of it.
NASA lists water, oxygen, and warmth as the three essential elements for life.
The Pajonales, an ancient lakebed on Earth that dates back four million years, is one of Mars’ closest analogues. Most life forms are thought to be unfriendly in this region. The high-altitude basin receives high UV radiation, hyper salinity, and low temperatures, which are comparable to the evaporitic basins of Mars.
Availability of Water
Using the model, the researchers gathered approximately 7700 photographs. The salt domes, boulders, and alabaster crystals that make up the surface of the basin were examined for the presence of any photosynthetic bacteria.
Regions were divided into four macro- and six micro-habitats using ground sampling data, 3D topographical mapping, and drone pictures. The team discovered that the microbial organisms were grouped in specific areas throughout the study site.
Dr. Freddie Kalaitzis from the Department of Computer Science at the University of Oxford claims that the model showed high predictive capability for the presence of geological materials that were very likely to contain biosignatures. The outcomes also matched up well with actual data.
Study of challenging ecosystems
The ability of the model to forecast the location of related but distinct natural systems is being put to the test by the researchers. The results of these studies will be used to test and refine hypotheses about how living things survive in harsh environments.
The study showed how machine learning techniques can speed up scientific discovery by analyzing huge amounts of diverse data and spotting patterns that are invisible to the human eye. The scientific community anticipates that the strategy will make it simpler to compile a database of biosignature probability and habitability algorithms, roadmaps, and models that can direct the investigation of life on Mars.
