A new analysis has cast doubt on the recent signal of phosphine gas in Venus’s atmosphere, a finding that could potentially be a sign of life. The scientists who originally discovered this in September found that the quantity of the gas seen couldn’t be produced by any known chemical processes on Venus.
Now, a study of the data behind the original paper suggests there are no signs of the gas after all.
The original study, led by Jane Greaves at Cardiff University in the UK, examined how light is absorbed as it passes through the Venusian atmosphere, leaving dark absorption lines in the light’s spectrum. Her team found an absorption line and identified it as phosphine.
Advertisement
Ignas Snellen at Leiden University in the Netherlands and his colleagues re-examined the data and found no such absorption line. The researchers say their new method of data analysis introduces fewer flaws.
This happens quite often in astronomy, with detections seen in objects that vanish when other people reduce the data, says Christopher Conselice at the University of Manchester in the UK, who wasn’t involved in either study.
Interferometry was used to gather the data. Instead of using one telescope, information is collected by an array of separate telescopes that then has to be pieced together. “[This] is probably one of the most complex types of astronomical data to analyse,” says Conselice.
Because of the complexity of the information, there are many ways to process it. It has to be reduced to remove disturbances or noise, and Snellan and his team say the original methods used to do this introduced errors, such as the phosphine signal. When they tried to replicate it, they found that five more absorption or emission signals had been spuriously added.
“The broader point raised by the new paper is an important one, regardless of how this all plays out,” says Brad Gibson at the University of Hull in the UK. Trying to draw conclusions when the dataset is so noisy is incredibly challenging, he says. “It demonstrates the fundamental challenge of working on important and exciting science when one is simultaneously working very near the limits of the data quality.”
While Snellan’s study is yet to be peer reviewed, some astronomers have said it is too early to speculate about what this means. Some are suggesting another independent analysis of the data is needed. Others argue we won’t have a conclusive answer until more data can be gathered. “Only new observations will be able to confirm the detection of this potentially biogenic gas,” says Abel Méndez at the University of Puerto Rico at Arecibo.
This comes after another analysis led by Clara Sousa-Silva at the Harvard-Smithsonian Center for Astrophysics in Massachusetts – who was also involved in the original phosphine observations – found no hint of phosphine on Venus when examining an older set of infrared data.
The original dataset used by both Greaves and Snellan’s teams has been removed from the  where all results from the Atacama Large Millimetre/submillimetre Array (ALMA) observatory are published because of a potential problem in the early stages of data processing. Researchers from Greaves’s team declined to comment until the new processing had been applied. “Until this process is completed, we cannot say whether the issue affected the detection of phosphine reported,” says an ALMA spokesperson.
“This is a great example of how science works,” says Conselice. “We test and deeply examine questions, especially important ones like this, and eventually get the right answer. Even if the process can be painful.”
Reference:
Sign up to our free Launchpad newsletter for a voyage across the galaxy and beyond, every Friday
