In the Bordeaux region of southwestern France, dozens of vineyard estates transform finicky grapes into bold red wine blends. Some bottles sell for thousands of dollars each. Prestigious chateaux boast the soil, microclimate and traditional methods that make their own wine superior, an unfathomable mix known as terroir.

“It’s one of those terms that the wine industry likes to keep a bit of mystery, part of the magic of wine,” says Alex Pouget, a computational neuroscientist at the University of Geneva.

Dr. Pouget tries to apply chemical precision to this je ne sais quoi. In a study Published Tuesday in the journal Communications Chemistry, he and his colleagues described a computer model that could identify which estate in Bordeaux produced a wine based solely on its chemical composition. The model also predicted the year the wine was made, known as the vintage, with an accuracy of about 50 percent.

Although wine connoisseurs often claim to be able to distinguish between wines from top estates, they rarely do blind taste tests, he said. “People have been making these claims for decades, but we’ve never actually had an objective measurement that showed this was true,” he said.

Dr. Pouget grew up in Paris in a family that drank only Bordeaux (“You act as if Burgundy doesn’t exist,” he said). As a young neuroscientist in the late 1980s, he studied the brain with machine learning, a form of artificial intelligence that identifies patterns in large data sets. He believed that these methods could be useful for the wine industry, but did not get around to testing the idea for thirty years.

He collaborated with Stéphanie Marchand from the Institute of Vine and Wine Science in Bordeaux, who had created a database of 80 wines of different vintages from seven castles. The database contained the chemical signatures of each wine, obtained from gas chromatography, an ancient and inexpensive method of breaking down substances into their molecular components.

The researchers trained an algorithm to look for common patterns in the wines’ chemical fingerprints. They were shocked by the results: the model grouped the wines into different clusters corresponding to their geographical location in the Bordeaux region. This showed that the particularities of each estate had drastically influenced the chemistry of the wines produced there, just as winemakers have claimed for centuries.

The estates gave the researchers permission to study their wines, on the condition that they were not mentioned by name. Dr. Pouget said that all the wines were part of the famous wines Bordeaux classification of 1855a ranking established by Napoleon III to promote the best wines of Bordeaux.

Dr. Pouget was surprised that the winemakers did not want to reveal their names because the study’s findings reinforced the idea that their wines were special. “I have scientific evidence that it makes sense to charge people for this because they produce something unique,” ​​he said, laughing.

Independent researchers said the study was part of a wave of recent research using machine learning to decipher terroir. “This is where the field is going and needs to go to make sense of the abundance of data,” says David Jeffery, an expert in wine chemistry at the University of Adelaide in Australia.

For example, he has used machine learning classify Shiraz wines from the Barossa Valley in Australia.

The approach, said Dr. Jeffery, is “not just about what makes a good wine chemically.” The models can also help producers adjust their growing and winemaking practices to preserve the character of their product in years of unexpected weather. “This is especially important in light of a changing climate,” he said.

Another application of these models, experts say, is dying out fraud, which is quite common with expensive wines. Producers have adapted their bottles, labels and corks to make them harder to copy.

“If there is any doubt about the origin of a wine, analyzing a wine from the estate as a benchmark would probably allow to know whether the wine is fake or not,” says Cornelis van Leeuwen, head of the viticulture and oenology department . at Bordeaux Sciences Agro.

The approach would likely work for any wine region, as long as the model is trained on a large number of wines from different producers and vintages, said Dr. van Leeuwen, who was not involved in the new research. An open question, however, is whether the model will maintain its accuracy after a few years, he said.

Dr. Pouget, who has a large wine collection, hopes to repeat the study with some of his favorite varieties from the Châteauneuf-du-Pape region in southeastern France.

But for the best wines, age is probably more important than provenance, he said.

“I only drink old wine,” he said. “I think drinking things when they’re under 15 is a bit criminal.”