Credit: HOWIE BLUESTEN

Serendipity is a term often misused, but if anyone in modern science could lay claim to it, it was Edward Lorenz. A meteorologist by training, he founded the entire discipline of chaos theory by accident — while attempting to improve a weather forecast.

Lorenz was born on 17 May 1917 in West Hartford, Connecticut, and recalled childhood interests in both mathematics and meteorology that seemed to predestine him for his later career. He first studied mathematics, but became a weather forecaster for the US Army Air Corps during the Second World War. In 1943, he took a masters degree in meteorology at the Massachusetts Institute of Technology, the beginning of a lifelong association with the institution. There, he set about using mathematical tools to break down complex climate phenomena, establishing a series of elegantly simple mathematical models to describe various aspects of atmospheric energetics and transport.

Lorenz was tinkering with one such model in 1961 when, wishing to repeat a simulation over a slightly longer timescale but disinclined to start over again, he began in the middle using values generated from the first run. The print-out with the input parameters rounded the six decimal places of the actual output to just three. Such a small discrepancy would not have been expected to materially affect the end result; but the second simulation produced an entirely different evolution from the first.

At first, Lorenz suspected a computer malfunction, but the irreproducibility of the result proved eminently reproducible. He was quick to grasp the implication that, as far as the atmosphere is concerned, “the prediction of the sufficiently distant future is impossible by any method, unless the present conditions are known exactly” (Lorenz, E. J. Atmos. Sci. 20, 130–141; 1963). In the absence of such knowledge, the question of what the weather will be in a month's time is one we are fated never to answer.

Lorenz first illustrated this with the metaphor that a flap of a seagull's wings was enough to alter the course of the weather forever. It was not until a decade later, in December 1972, at an invited talk at a meeting of the American Association for the Advancement of Science, that he used the description that was to lodge chaos in the public imagination: “Does the flap of a butterfly's wings in Brazil set off a tornado in Texas?” That vividly framed question had a perfect visual accompaniment in tracings of the 'Lorenz attractor', a three-dimensional representation of chaotic flow established by Lorenz when considering atmospheric convection, whose paths resemble nothing more than a pair of butterfly wings.

For establishing the theoretical basis of climate predictability, Lorenz was awarded the Crafoord prize by the Royal Swedish Academy of Sciences in 1983, and the Kyoto prize of the Inamori foundation in 1991. The citation for the latter prize attested Lorenz as having “brought about one of the most dramatic changes in mankind's view of nature since Sir Isaac Newton”. Edward Lorenz died on 16 April at his home in Cambridge, Massachusetts, at the age of 90.