A simple weather model rerun in 1963 changed what scientists thought prediction itself could do

In the early 1960s, meteorologist Edward Lorenz reran his weather model but with some figures rounded off to make the calculations easier. What happened next became one of the most important events in the world of modern science. In the second forecast, although nearly all the numbers started out identical to those in the first run, there was a dramatic deviation from the initial results. This unpredicted variation forced scientists to question their assumptions about predictability.This case has since been hailed as the birth of chaos theory. Indeed, the MIT Lorenz Center says that Lorenz’s work showed the world that a system of equations could produce unpredictable results.The rerun that changed forecastingThe well-known example took place when Lorenz was conducting experiments with a pioneering computer program that simulated the behaviour of the atmosphere. According to Physics Today, Lorenz had been experimenting with a simplified dynamical weather system on a Royal McBee LGP-30 computer.One day, when re-running the simulation, Lorenz used rounded numbers instead of the longer decimals for his initial calculations. It may have seemed like a small detail. However, the simulation's outcome began to diverge from the previous one very quickly.Instead of considering the divergence as simply a mistake made by the machine, Lorenz saw a larger principle emerging. Small alterations in the initial conditions caused completely different outcomes. Something which seemed to be an attempt to simplify the calculations turned out to be a fundamental constraint on prediction.As stated by MIT News, Lorenz realised that “small differences in a dynamic system such as the atmosphere could trigger vast and often unsuspected results”.Why the discovery matteredPrior to Lorenz's work, most researchers believed that improvements in data acquisition and computing capabilities would enable highly accurate long-range weather predictions. The findings of Lorenz contradicted that confidence.The issue was not simply that weather systems were complicated. Lorenz showed that some systems are sensitive in a very specific way. Lorenz found that certain systems are characterised by sensitivities that are quite particular. An error as small as one part in ten million in the data collection could lead to inaccurate predictions within weeks.According to an article, Lorenz's 1963 paper marked the shift in thinking about chaos, attractors, sensitivity to initial conditions, and the limits to forecasting of nonlinear systems. This new understanding affected the philosophical approach to forecasting. Scientists began to realise that there was a certain inevitability to unpredictability. From weather model to chaos theoryLorenz’s finding did not stop there; his further research led to the creation of the now well-known Lorenz attractor, which is a mathematical representation of chaotic motion that remains one of the most recognisable images in modern science.Reportedly, the Lorenz equations showed that even when the system follows a precise set of laws, small differences lead to unpredictable behaviour in the long run. This formed the basis of the emerging field of chaos theory, whereby researchers try to study complex systems that seem orderly yet are very sensitive to any small disturbances.Reports indicated that the findings of Lorenz eventually impacted other areas of study, including math, physics, biology, and climatology, among others.A lasting impact on weather scienceAlmost six decades later, the discovery by Lorenz continues to inform modern-day forecasting. The models have grown extremely complex, but science still operates within the confines of predictability outlined by Lorenz. The discovery of Lorenz led to the move from perfect deterministic forecasts in meteorology to more realistic probabilities. Predictions today consider probabilities and ranges of reliability rather than perfect prediction.The interaction of chaos with climate systems is also a topic for ongoing research. As short-term weather predictions remain confined, climate science examines broader trends that can still be statistically analysed. The tale endures partly because it translates an abstract scientific concept into something tangible. By rounding off some figures on his computer, a scientist stumbled upon the phenomenon where nature amplifies minute differences into major events.The rerun experiment by Lorenz did not mark the end of forecasting. Rather, it taught researchers about what could reasonably be accomplished by predictions. The lesson was both humbling and transformative: systems may obey fixed laws, yet still resist precise long-range prediction.This little numeric trick turned out to be one of the greatest discoveries of twentieth-century science.