El Niños are among the most influential events affecting weather conditions, driving devastating droughts in some regions and floods elsewhere that can kill hundreds of thousands combined. They also seem to be becoming more common, but only now do we have the long-term data to confirm this.

Every credible climate model agrees that extra greenhouse gasses warm the world. However, there are other matters where their results conflict. “Most models predict an increase in El Niño frequency with global warming, but not all agree,” University of Melbourne PhD student and lead author Mandy Freund told IFLScience. Adding to the confusion, climate scientists now recognize two types of El Ninos, beginning in the Eastern and Central Pacific respectively, that can affect many regions differently.

Confronted with this disagreement, the obvious thing to do is to see whether the last 30 years have had more El Niños than normal. Unfortunately, we lack the long-term records to know what normal is. Some weather events can be traced through their effects on tree growth, but it is difficult to identify past El Niños in this way and nearly impossible to distinguish the Central Pacific events from Eastern ones. The Pacific Ocean has no trees and, as Freund explained to IFLScience, nearby tropical forests most affected by these events don't have species that reveal seasonal variations through tree rings well.

So Freund turned to the Pacific's great resource: coral. Corals have growth rings like trees, with the isotopes deposited in them revealing information about sea temperatures and salinity in the area where they grow. Using data collected from coral cores at 27 sites around the great ocean, she trained an algorithm to recognize patterns that not only signify El Niños but to distinguish between the types based on the locations where anomalies appear first.

Once the algorithm could identify known El Niños, it was turned loose on earlier data to find earlier events. The cores Freund relied on go back 400 years. Although even older cores exist, Freund regards the coverage as too patchy to be reliable.

Freund's approach was considered unworkable by many scientists since the two El Niño types produce such subtle differences in coral composition. She had to work hard to get a team of climate and coral scientists on board to support her giving it a try.

Not only was Freund able to distinguish El Niños by type but she found that the last 30 years were unprecedented, with Central Pacific El Niños almost tripling in frequency, while fewer but more intense Eastern Pacific events occurred, Freund reports in Nature Geoscience.

"By understanding the past, we are better equipped to understand the future, especially in the context of climate change," Freund said in a statement.