Dr. Edward Norton Lorenz, the American mathematician and meteorologist who revolutionized weather forecasting through his discovery of the Chaos Theory, recently died at the age of 90. Lorenz was attempting to predict the weather through computers but ended up fathering the modern field of "deterministic chaos."
Around 1960, he constructed a mathematical model demonstrating the way air moves around in the atmosphere. However, while studying weather patterns, he came to the conclusion that they did not always behave as it was predicted they would.
He found that just minute variations in the initial values of variables in his 12-variable computer weather model would be enough to produce grossly different patterns.
The sensitive dependence on initial conditions not altering even slightly became known as the "butterfly effect" — the idea being that a small disturbance like the flapping of a butterfly's wings can spur enormous consequences.
Lorenz was running weather simulations on the computer model and one day in the winter of 1961, decided to repeat the simulations for a longer time. But instead of repeating the entire simulation, he began the second run in the middle, typing in numbers from the first run for the initial conditions.
Since the computer program was the same, he surmised the weather patterns for the second run should exactly duplicate those of the first. But instead, the two weather system trajectories promptly took off in two separate paths.
Lorenz said later that he was so surprised by this development that he assumed his computer must have been malfunctioning. But after determining that was not the case, he typed in new conditions, this time entering rounded-off numbers.
However, even this small discrepancy of less than 0.1 per cent, produced a different end result.
Lorenz explored the underlying mathematics and eventually published his findings in a paper entitled Deterministic Nonperiodic Flow. In the paper, he espoused the Chaos Theory regarding weather forecasting — namely that perfect weather prediction was impossible.
A perfect forecast would require not only a perfect computer model, but also perfect knowledge of the temperature, humidity and other weather conditions anywhere around the globe at one moment of time. The slightest discrepancy could mean a different outcome.