# Edward Lorenz The American meteorologist and mathematician who **founded modern chaos theory** through his accidental 1961 discovery that small rounding differences in weather-simulation initial conditions produced dramatically different long-term outputs. Lorenz (1917–2008) was an MIT atmospheric scientist running early computer simulations of weather; his 1963 paper "Deterministic Nonperiodic Flow" established the mathematical foundations of what became chaos theory. His 1972 lecture titled "Does the Flap of a Butterfly's Wings in Brazil Set Off a Tornado in Texas?" gave the field its enduring metaphor. ## Career and contribution **Background.** Dartmouth (BA 1938), Harvard (MA 1940), MIT (ScD 1948). Wartime weather-forecaster for the Army Air Corps. Returned to MIT after WWII, where he spent the rest of his career. **The 1961 discovery.** Running a weather-simulation model on an LGP-30 computer in 1961, Lorenz wanted to continue a previous simulation from an intermediate state. He typed in the saved values from a printout, which had rounded to three decimal places, while the original computation used six. The resulting weather pattern diverged dramatically from the original. The rounding error — at the fourth decimal place — produced completely different outputs over a few simulated weeks. The implications: real weather systems (and other nonlinear systems) are sensitively dependent on initial conditions; arbitrarily small differences in inputs produce arbitrarily large differences in outputs over time; prediction beyond a horizon is structurally impossible regardless of measurement precision. **The 1963 paper.** "Deterministic Nonperiodic Flow" (*Journal of the Atmospheric Sciences*, 1963) established the mathematical framework. The "Lorenz attractor" (the strange attractor produced by Lorenz's simplified weather equations) became one of the iconic images of chaos theory. **The butterfly metaphor.** Lorenz's 1972 American Association for the Advancement of Science talk titled "Does the Flap of a Butterfly's Wings in Brazil Set Off a Tornado in Texas?" gave the field its popular name. The literal claim about butterflies and tornadoes is metaphorical; the underlying point about sensitive dependence on initial conditions is mathematically rigorous. ## KB significance **Why long-range prediction is structurally limited.** Lorenz's work establishes that prediction of weather (and many other complex systems) has a fundamental horizon beyond which forecasts become unreliable regardless of how good the model is. The political application: long-range political forecasting is structurally limited in similar ways. Election forecasts beyond 6-12 months become unreliable not because forecasters are inadequate but because the system has chaotic dynamics. **The Mandelbrot connection.** Lorenz and Benoit Mandelbrot worked on related problems in different domains (Lorenz on weather; Mandelbrot on fractals and finance). Their frameworks together established the mathematical machinery for understanding how complex systems produce extreme outputs from small inputs — the foundation of contemporary thinking about black swans, fat tails, and tail-risk. ## See also - [[Chaos Theory]] — the field Lorenz founded - [[Butterfly Effect]] — the eponymous concept - [[Benoit Mandelbrot]] — contemporary mathematician working on adjacent problems - [[Fractals]] — related framework for nonlinear dynamics