Every June, the sky over India fills with moisture, and how much falls determines rice harvests, water reserves, and food prices for a billion people. But predicting the monsoon's strength in advance was like guessing someone's mood from a single photo: too many hidden springs. Official forecasts often missed the mark, and the connection with El Niño, relied on for decades, has weakened in recent years. Now climatologists have found an unexpected oracle: a lone point in the South Atlantic, far from Indian shores.

The researchers took surface temperature data from 1948 to 2019 and, for each pair of points, calculated how synchronously they warmed and cooled, with delays up to 200 days. The strongest links, the top 5 percent, formed a directed network. Then, for each point, they counted how many links entered it and found the one whose 'incoming popularity' best predicts the next monsoon's rainfall. It turned out to be a point near 30 degrees south latitude and 30 degrees west longitude, in the open ocean.

Testing on an independent period from 1989 to 2018 gave a correlation of 0.5 between predicted and actual rainfall. That is not perfect, but much better than before: official forecasts often erred more, and the best models only achieved such accuracy a month before the season. For 2019, the forecast differed from reality by less than 2 percent. For 2020, the method predicted 842.85 millimeters of rainfall, which is 5 percent below normal.

Now farmers and authorities can plan planting and water reserves not a month but five months ahead. The method requires no supercomputers or complex models; just watching one point in the ocean. But it is not a magic button: the mechanism behind the link still needs to be understood, and warming will continue to reshape the network, so the 'oracle' will need regular recalibration. The pulse of the global climate, felt at a single point, already helps us hear the breath of the coming monsoon.