El Niño's Past 21000 Years Predicts Changes in Future Climate

El Niño, the weather system that can cause massive changes in climate, is important to understand when it comes to predicting how it influences future conditions. Now, scientists have taken a closer look at El Niño's past in order to predict how it might respond to global warming.

In this case, researchers see a large amount of variability in the El Niño-Southern Oscillation (ENSO) when looking back at climate records from thousands of years ago. Without a clear understanding of what caused past changes in ENSO variability, though, predicting its future is difficult.

In order to learn a bit more about El Niño, the researchers analyzed a series of transient Coupled General Circulation Model simulations forced by changes in greenhouse gases, orbital forcing, meltwater discharge and the ice-sheet history over the past 21,000 years. This is actually the farthest in the past that this model has ever been run continuously, and took months in order to simulate.

"The model gives some very clear predictions that are very much in line with some of the best understandings of the physics controlling the El Niño system," said Kim Cobb, one of the researchers, in a news release. "It shows that this climate system in the model is sensitive to a variety of different natural climate changes that occurred over the last 21,000 years."

More specifically, the scientists found that during the interglacial period, ENSO strengthened due to increasing positive ocean-atmosphere feedbacks. They also found that ENSO characteristics change drastically in response to meltwater discharges during early deglaciation, and that increasing deglacial atmospheric carbon dioxide concentrations tend to weaken ENSO.  Yet melting ice sheets intensify ENSO.

"The more we can close the loop between what this model says happened in the past and what the data set happened in the past, then we can project forward our improved understanding to understand future El Niño," said Cobb.

The findings are published in the journal Nature.