Chaos theory examines how deterministic systems exhibit unpredictable, sensitive dependence on initial conditions.Chaos is not synonymous with disorder. It is orderly unpredictability. Look at the smoke rising from a cigarette: at first it flows in a smooth stream, but higher up it suddenly turns into billowing curls. The transition from laminar to turbulent flow is a typical scenario for the birth of chaos. The equations describing such processes are deterministic (containing no randomness), but so sensitive to initial data that the slightest rounding error grows and nullifies any forecast. This property is called exponential divergence of trajectories.
Although the first intuitive insights about the instability of certain motions were expressed by Henri Poincaré as early as the 1890s, modern chaos theory was born in the 1960s–1970s. Edward Lorenz of the Massachusetts Institute of Technology discovered chaos in a simplified model of atmospheric convection in 1963. Mathematician James Yorke and his student Tien-Yien Li introduced the term 'chaos' in 1975. Mitchell Feigenbaum in 1978 discovered universal patterns in the transition to chaos through a period-doubling cascade, linking them to constants named after him.
The starting point is often the Lorenz system: three ordinary differential equations with two nonlinear terms of the xy and xz type. At certain parameter values (σ=10, ρ=28, β=8/3), the phase trajectories form a chaotic attractor—they endlessly wind around two regions, never intersecting or looping. The evolution of such a system has the property of mixing: any, even tiny, initial cloud of points eventually spreads uniformly over the attractor. Information about the initial state is irretrievably lost, although the equations themselves are reversible. This leads to the practical impossibility of long-term forecasting, despite determinism. Statistical methods, such as constructing return maps (Poincaré sections) and calculating the power spectrum, help distinguish chaos from periodic regimes and noise.
Weather forecasts are inaccurate beyond two weeks due to the chaotic nature of the atmosphere, and encryption algorithms use chaos to protect data.
In 1961, Edward Lorenz, restarting a long computer weather calculation, entered 0.506 instead of 0.506127—and got a completely different climate. Thus was born the understanding that predictability even in simple systems is limited to a horizon of a few characteristic times.












