Prediction is very difficult, especially if it is about the future, the great physicist Niels Bohr once observed. Eric Fischer, a climate scientist at ETH-Zurich, has just arrived at much the same conclusion.
Fischer and colleagues reported in Nature Climate Change that their attempts to look at weather patterns even in the near future delivered very large uncertainties. This, they report, is a consequence of something sometimes called the butterfly’s wing effect and sometimes chaos theory.
Meteorologists are fond of saying that the flap of a butterfly’s wing in Brazil could set up a chain of air movements that might end with a tornado in Texas. Even the tiniest change in the initial starting conditions can produce wildly different outcomes later.
Fischer and his fellow researchers simply wanted to solve a puzzle for the farmers, decision-makers and civic authorities who must prepare for floods, heat waves and ice storms in the next decade or two, and who have already been urged to face up to the need to adapt to climate change.
The Zurich team used a climate model to simulate extremes of temperature and precipitation between the periods 2016-2035 and 2041-2060. They matched the model with real, historical data from 1950 to 2005, they assumed that greenhouse gas emissions would be the same in all simulations, and they ran the model 21 times, each time making a tiny random variation in the global atmospheric temperature on the first day: 1 January 1950.
