In 1963, Edward Lorenz Re-Ran a Weather Model: A Tiny Change Rewrote the Forecast and Revealed Chaos Theory

An event occurred in 1963 when Edward Lorenz was simulating the weather process at MIT. He repeated the process with the same data but rounded the numbers to save himself some time. The expectation was that he should obtain similar results to those previously, but to his astonishment, he obtained quite different results. This event became one of the greatest events in the science of the 20th century. Lorenz did not make any mistakes either with computations or algorithms. His simulation was a deterministic model whose results depended only on the calculations. Nonetheless, slight differences in the initial conditions resulted in completely different output results.

This fact gave rise to the revolutionary idea about the limitations of the predictability of a process in science. Before that, scientists believed that precise forecasting could be achieved if enough data and advanced technologies were used. However, Lorenz showed with his experiment that all systems are unpredictable due to their internal inconsistencies.

Why Did a Small Number Make a Big Difference?

This is what led Lorenz to his discovery: the difference came the moment he made the decision to round off numbers to just three digits instead of six. This minute change in the values created a significant difference over time. As further studies carried out by the MIT Lorenz Center and the National Oceanic and Atmospheric Administration proved, this was one of the properties of chaos, not something connected to the modeling. It has become obvious that perfection would not guarantee the best results, and scientists would need to deal with increasing uncertainty, particularly in weather predictions, because of the high number of variables.

The term "butterfly effect" was used in the context of Lorenz's speech delivered in 1972. However, the theory was developed much earlier in 1963. The theory says that there are no cases where a butterfly causes a hurricane to occur. Instead, Lorenz's theory says that there could be some small variations affecting the outcome. A study conducted by NOAA makes a clear distinction between the concept of the butterfly and Lorenz's theory, saying that initially, there were no links between them.

Forecasting Could Never Be Perfect

Thus, meteorologists had to reconsider their objective. Rather than pursuing accuracy over a lengthy period of time, they needed to concentrate on probability and uncertainty. Studies conducted at MIT found that accurate weather predictions can last only for two or three weeks at best, irrespective of any technological advancements. Such an outcome should not be regarded as an inadequacy of science, as it was only the result of a better understanding of natural phenomena. Although the atmosphere operates under physical laws, its interaction makes it impossible to forecast anything accurately after a few weeks. Such ideas affected other sciences that are based on complex systems, such as climatology, economics, and engineering.

Chaos theory is used today by researchers at NASA and the National Academies to explore different Earth systems. Lorenz's work became the basis for the development of large system evolution studies when small changes cause big differences.