Chaos theory

From Wikipedia, the free encyclopedia

Jump to navigationJump to search
For other uses, see Chaos theory (disambiguation) and Chaos (disambiguation).

A plot of the Lorenz attractor for values r = 28, σ = 10, b = 8/3

An animation of a double-rod pendulum at an intermediate energy showing chaotic

behavior. Starting the pendulum from a slightly different initial condition would
result in a vastly different trajectory. The double-rod pendulum is one of the
simplest dynamical systems with chaotic solutions.
Chaos theory is an interdisciplinary theory and branch of mathematics focusing on
the study of chaos: dynamical systems whose apparently random states of disorder
and irregularities are actually governed by underlying patterns and deterministic
laws that are highly sensitive to initial conditions.[1][2] Chaos theory states
that within the apparent randomness of chaotic complex systems, there are
underlying patterns, interconnectedness, constant feedback loops, repetition, self-
similarity, fractals, and self-organization.[3] The butterfly effect, an underlying
principle of chaos, describes how a small change in one state of a deterministic
nonlinear system can result in large differences in a later state (meaning that
there is sensitive dependence on initial conditions).[4] A metaphor for this
behavior is that a butterfly flapping its wings in Brazil can cause a tornado in
Texas.[5]

Small differences in initial conditions, such as those due to errors in

measurements or due to rounding errors in numerical computation, can yield widely
diverging outcomes for such dynamical systems, rendering long-term prediction of
their behavior impossible in general.[6] This can happen even though these systems
are deterministic, meaning that their future behavior follows a unique evolution[7]
and is fully determined by their initial conditions, with no random elements
involved.[8] In other words, the deterministic nature of these systems does not
make them predictable.[9][10] This behavior is known as deterministic chaos, or
simply chaos. The theory was summarized by Edward Lorenz as:[11]

Chaos: When the present determines the future, but the approximate present does not
approximately determine the future.

Chaotic behavior exists in many natural systems, including fluid flow, heartbeat
irregularities, weather and climate.[12][13][7] It also occurs spontaneously in
some systems with artificial components, such as the stock market and road traffic.
[14][3] This behavior can be studied through the analysis of a chaotic mathematical
model, or through analytical techniques such as recurrence plots and Poincaré maps.
Chaos theory has applications in a variety of disciplines, including meteorology,
[7] anthropology,[15] sociology, environmental science, computer science,
engineering, economics, ecology, pandemic crisis management,[16][17]. The theory
formed the basis for such fields of study as complex dynamical systems, edge of
chaos theory, and self-assembly processes.

Contents
1 Introduction
2 Chaotic dynamics
2.1 Chaos as a spontaneous breakdown of topological supersymmetry
2.2 Sensitivity to initial conditions
2.3 Non-periodicity
2.4 Topological mixing
2.5 Topological transitivity
2.6 Density of periodic orbits
2.7 Strange attractors
2.8 Minimum complexity of a chaotic system
2.9 Infinite dimensional maps
2.10 Jerk systems
3 Spontaneous order
4 History
5 Applications
5.1 Cryptography
5.2 Robotics
5.3 Biology
5.4 Economics
5.5 Other areas
6 See also
7 References
8 Further reading
8.1 Articles
8.2 Textbooks
8.3 Semitechnical and popular works
9 External links
Introduction
Chaos theory concerns deterministic systems whose behavior can, in principle, be
predicted. Chaotic systems are predictable for a while and then 'appear' to become
random. The amount of time that the behavior of a chaotic system can be effectively
predicted depends on three things: how much uncertainty can be tolerated in the
forecast, how accurately its current state can be measured, and a time scale
depending on the dynamics of the system, called the Lyapunov time. Some examples of
Lyapunov times are: chaotic electrical circuits, about 1 millisecond; weather
systems, a few days (unproven); the inner solar system, 4 to 5 million years.[18]
In chaotic systems, the uncertainty in a forecast increases exponentially with
elapsed time. Hence, mathematically, doubling the forecast time more than squares
the proportional uncertainty in the forecast. This means, in practice, a meaningful
prediction cannot be made over an interval of more than two or three times the
Lyapunov time. When meaningful predictions cannot be made, the system appears
random.[19]

Chaos theory is a method of qualitative and quantitative analysis to investigate

the behavior of dynamic systems that cannot be explained and predicted by single
data relationships, but must be explained and predicted by whole, continuous data
relationships.

Chaotic dynamics

The map defined by x → 4 x (1 – x) and y → (x + y) mod 1 displays sensitivity to

initial x positions. Here, two series of x and y values diverge markedly over time
from a tiny initial difference.
In common usage, "chaos" means "a state of disorder".[20][21] However, in chaos
theory, the term is defined more precisely. Although no universally accepted
mathematical definition of chaos exists, a commonly used definition, originally
formulated by Robert L. Devaney, says that to classify a dynamical system as
chaotic, it must have these properties:[22]

it must be sensitive to initial conditions,

it must be topologically transitive,
it must have dense periodic orbits.
In some cases, the last two properties above have been shown to actually imply
sensitivity to initial conditions.[23][24] In the discrete-time case, this is true
for all continuous[disambiguation needed] maps on metric spaces.[25] In these
cases, while it is often the most practically significant property, "sensitivity to
initial conditions" need not be stated in the definition.
If attention is restricted to intervals, the second property implies the other two.
[26] An alternative and a generally weaker definition of chaos uses only the first
two properties in the above list.[27]

Chaos as a spontaneous breakdown of topological supersymmetry

Main article: Supersymmetric theory of stochastic dynamics
In continuous time dynamical systems, chaos is the phenomenon of the spontaneous
breakdown of topological supersymmetry, which is an intrinsic property of evolution
operators of all stochastic and deterministic (partial) differential equations.[28]
[29] This picture of dynamical chaos works not only for deterministic models, but
also for models with external noise which is an important generalization from the
physical point of view, since in reality, all dynamical systems experience
influence from their stochastic environments. Within this picture, the long-range
dynamical behavior associated with chaotic dynamics (e.g., the butterfly effect) is
a consequence of Goldstone's theorem—in the application to the spontaneous
topological supersymmetry breaking.

Sensitivity to initial conditions

Main article: Butterfly effect

Lorenz equations used to generate plots for the y variable. The initial conditions
for x and z were kept the same but those for y were changed between 1.001, 1.0001
and 1.00001. The values for {\displaystyle \rho }\rho , {\displaystyle
\sigma }\sigma and {\displaystyle \beta }\beta were 45.92, 16 and 4 respectively.
As can be seen from the graph, even the slightest difference in initial values
causes significant changes after about 12 seconds of evolution in the three cases.
This is an example of sensitive dependence on initial conditions.
Sensitivity to initial conditions means that each point in a chaotic system is
arbitrarily closely approximated by other points that have significantly different
future paths or trajectories. Thus, an arbitrarily small change or perturbation of
the current trajectory may lead to significantly different future behavior.[3]

Sensitivity to initial conditions is popularly known as the "butterfly effect", so-

called because of the title of a paper given by Edward Lorenz in 1972 to the
American Association for the Advancement of Science in Washington, D.C., entitled
Predictability: Does the Flap of a Butterfly's Wings in Brazil set off a Tornado in
Texas?.[30] The flapping wing represents a small change in the initial condition of
the system, which causes a chain of events that prevents the predictability of
large-scale phenomena. Had the butterfly not flapped its wings, the trajectory of
the overall system could have been vastly different.

A consequence of sensitivity to initial conditions is that if we start with a

limited amount of information about the system (as is usually the case in
practice), then beyond a certain time, the system would no longer be predictable.
This is most prevalent in the case of weather, which is generally predictable only
about a week ahead.[31] This does not mean that one cannot assert anything about
events far in the future—only that some restrictions on the system are present. For
example, we do know with weather that the temperature will not naturally reach 100
°C or fall to −130 °C on earth (during the current geologic era), but that does not
mean that we can predict exactly which day will have the hottest temperature of the
year.

In more mathematical terms, the Lyapunov exponent measures the sensitivity to

initial conditions, in the form of rate of exponential divergence from the
perturbed initial conditions.[32] More specifically, given two starting
trajectories in the phase space that are infinitesimally close, with initial
separation {\displaystyle \delta \mathbf {Z} _{0}}\delta \mathbf {Z} _{0}, the two
trajectories end up diverging at a rate given by
{\displaystyle |\delta \mathbf {Z} (t)|\approx e^{\lambda t}|\delta \mathbf {Z}
_{0}|,}{\displaystyle |\delta \mathbf {Z} (t)|\approx e^{\lambda t}|\delta \mathbf
{Z} _{0}|,}
where {\displaystyle t}t is the time and {\displaystyle \lambda }\lambda is the
Lyapunov exponent. The rate of separation depends on the orientation of the initial
separation vector, so a whole spectrum of Lyapunov exponents can exist. The number
of Lyapunov exponents is equal to the number of dimensions of the phase space,
though it is common to just refer to the largest one. For example, the maximal
Lyapunov exponent (MLE) is most often used, because it determines the overall
predictability of the system. A positive MLE is usually taken as an indication that
the system is chaotic.[7]

In addition to the above property, other properties related to sensitivity of

initial conditions also exist. These include, for example, measure-theoretical
mixing (as discussed in ergodic theory) and properties of a K-system.[10]

Non-periodicity
A chaotic system may have sequences of values for the evolving variable that
exactly repeat themselves, giving periodic behavior starting from any point in that
sequence. However, such periodic sequences are repelling rather than attracting,
meaning that if the evolving variable is outside the sequence, however close, it
will not enter the sequence and in fact, will diverge from it. Thus for almost all
initial conditions, the variable evolves chaotically with non-periodic behavior.

Topological mixing

Six iterations of a set of states {\displaystyle [x,y]}[x,y] passed through the

logistic map. The first iterate (blue) is the initial condition, which essentially
forms a circle. Animation shows the first to the sixth iteration of the circular
initial conditions. It can be seen that mixing occurs as we progress in iterations.
The sixth iteration shows that the points are almost completely scattered in the
phase space. Had we progressed further in iterations, the mixing would have been
homogeneous and irreversible. The logistic map has equation {\displaystyle
x_{k+1}=4x_{k}(1-x_{k})}{\displaystyle x_{k+1}=4x_{k}(1-x_{k})}. To expand the
state-space of the logistic map into two dimensions, a second state, {\displaystyle
y}y, was created as {\displaystyle y_{k+1}=x_{k}+y_{k}}{\displaystyle y_{k+1}=x_{k}
+y_{k}}, if {\displaystyle x_{k}+y_{k}<1}{\displaystyle x_{k}+y_{k}<1} and
{\displaystyle y_{k+1}=x_{k}+y_{k}-1}{\displaystyle y_{k+1}=x_{k}+y_{k}-1}
otherwise.

The map defined by x → 4 x (1 – x) and y → (x + y) mod 1 also displays topological

mixing. Here, the blue region is transformed by the dynamics first to the purple
region, then to the pink and red regions, and eventually to a cloud of vertical
lines scattered across the space.
Topological mixing (or the weaker condition of topological transitivity) means that
the system evolves over time so that any given region or open set of its phase
space eventually overlaps with any other given region. This mathematical concept of
"mixing" corresponds to the standard intuition, and the mixing of colored dyes or
fluids is an example of a chaotic system.

Topological mixing is often omitted from popular accounts of chaos, which equate
chaos with only sensitivity to initial conditions. However, sensitive dependence on
initial conditions alone does not give chaos. For example, consider the simple
dynamical system produced by repeatedly doubling an initial value. This system has
sensitive dependence on initial conditions everywhere, since any pair of nearby
points eventually becomes widely separated. However, this example has no
topological mixing, and therefore has no chaos. Indeed, it has extremely simple
behavior: all points except 0 tend to positive or negative infinity.
Topological transitivity
A map {\displaystyle f:X\to X}{\displaystyle f:X\to X} is said to be topologically
transitive if for any pair of non-empty open sets {\displaystyle U,V\subset X}
{\displaystyle U,V\subset X}, there exists {\displaystyle k>0}k>0 such that
{\displaystyle f^{k}(U)\cap V\neq \emptyset }{\displaystyle f^{k}(U)\cap V\neq
\emptyset }. Topological transitivity is a weaker version of topological mixing.
Intuitively, if a map is topologically transitive then given a point x and a region
V, there exists a point y near x whose orbit passes through V. This implies that is
impossible to decompose the system into two open sets.[33]

An important related theorem is the Birkhoff Transitivity Theorem. It is easy to

see that the existence of a dense orbit implies in topological transitivity. The
Birkhoff Transitivity Theorem states that if X is a second countable, complete
metric space, then topological transitivity implies the existence of a dense set of
points in X that have dense orbits.[34]

Density of periodic orbits

For a chaotic system to have dense periodic orbits means that every point in the
space is approached arbitrarily closely by periodic orbits.[33] The one-dimensional
logistic map defined by x → 4 x (1 – x) is one of the simplest systems with density
of periodic orbits. For example, {\displaystyle {\tfrac {5-{\sqrt {5}}}{8}}}{\tfrac
{5-{\sqrt {5}}}{8}} → {\displaystyle {\tfrac {5+{\sqrt {5}}}{8}}}{\tfrac {5+{\sqrt
{5}}}{8}} → {\displaystyle {\tfrac {5-{\sqrt {5}}}{8}}}{\tfrac {5-{\sqrt {5}}}{8}}
(or approximately 0.3454915 → 0.9045085 → 0.3454915) is an (unstable) orbit of
period 2, and similar orbits exist for periods 4, 8, 16, etc. (indeed, for all the
periods specified by Sharkovskii's theorem).[35]

Sharkovskii's theorem is the basis of the Li and Yorke[36] (1975) proof that any
continuous one-dimensional system that exhibits a regular cycle of period three
will also display regular cycles of every other length, as well as completely
chaotic orbits.

Strange attractors

The Lorenz attractor displays chaotic behavior. These two plots demonstrate
sensitive dependence on initial conditions within the region of phase space
occupied by the attractor.
Some dynamical systems, like the one-dimensional logistic map defined by x → 4 x (1
– x), are chaotic everywhere, but in many cases chaotic behavior is found only in a
subset of phase space. The cases of most interest arise when the chaotic behavior
takes place on an attractor, since then a large set of initial conditions leads to
orbits that converge to this chaotic region.[37]

An easy way to visualize a chaotic attractor is to start with a point in the basin
of attraction of the attractor, and then simply plot its subsequent orbit. Because
of the topological transitivity condition, this is likely to produce a picture of
the entire final attractor, and indeed both orbits shown in the figure on the right
give a picture of the general shape of the Lorenz attractor. This attractor results
from a simple three-dimensional model of the Lorenz weather system. The Lorenz
attractor is perhaps one of the best-known chaotic system diagrams, probably
because it is not only one of the first, but it is also one of the most complex,
and as such gives rise to a very interesting pattern that, with a little
imagination, looks like the wings of a butterfly.

Unlike fixed-point attractors and limit cycles, the attractors that arise from
chaotic systems, known as strange attractors, have great detail and complexity.
Strange attractors occur in both continuous dynamical systems (such as the Lorenz
system) and in some discrete systems (such as the Hénon map). Other discrete
dynamical systems have a repelling structure called a Julia set, which forms at the
boundary between basins of attraction of fixed points. Julia sets can be thought of
as strange repellers. Both strange attractors and Julia sets typically have a
fractal structure, and the fractal dimension can be calculated for them.

Minimum complexity of a chaotic system

Bifurcation diagram of the logistic map x → r x (1 – x). Each vertical slice shows
the attractor for a specific value of r. The diagram displays period-doubling as r
increases, eventually producing chaos.
Discrete chaotic systems, such as the logistic map, can exhibit strange attractors
whatever their dimensionality. Universality of one-dimensional maps with parabolic
maxima and Feigenbaum constants {\displaystyle \delta =4.669201...}
{\displaystyle \delta =4.669201...},{\displaystyle \alpha =2.502907...}
{\displaystyle \alpha =2.502907...}[38][39] is well visible with map proposed as a
toy model for discrete laser dynamics: {\displaystyle x\rightarrow Gx(1-\mathrm
{tanh} (x))}{\displaystyle x\rightarrow Gx(1-\mathrm {tanh} (x))}, where
{\displaystyle x}x stands for electric field amplitude, {\displaystyle G}G[40] is
laser gain as bifurcation parameter. The gradual increase of {\displaystyle G}G at
interval {\displaystyle [0,\infty )}[0,\infty ) changes dynamics from regular to
chaotic one[41] with qualitatively the same bifurcation diagram as those for
logistic map.

In contrast, for continuous dynamical systems, the Poincaré–Bendixson theorem shows

that a strange attractor can only arise in three or more dimensions. Finite-
dimensional linear systems are never chaotic; for a dynamical system to display
chaotic behavior, it must be either nonlinear or infinite-dimensional.

The Poincaré–Bendixson theorem states that a two-dimensional differential equation

has very regular behavior. The Lorenz attractor discussed below is generated by a
system of three differential equations such as:

{\displaystyle {\begin{aligned}{\frac {\mathrm {d} x}{\mathrm {d} t}}&=\sigma

y-\sigma x,\\{\frac {\mathrm {d} y}{\mathrm {d} t}}&=\rho x-xz-y,\\{\frac {\mathrm
{d} z}{\mathrm {d} t}}&=xy-\beta z.\end{aligned}}}{\begin{aligned}{\frac {\mathrm
{d} x}{\mathrm {d} t}}&=\sigma y-\sigma x,\\{\frac {\mathrm {d} y}{\mathrm {d}
t}}&=\rho x-xz-y,\\{\frac {\mathrm {d} z}{\mathrm {d} t}}&=xy-\beta
z.\end{aligned}}
where {\displaystyle x}x, {\displaystyle y}y, and {\displaystyle z}z make up the
system state, {\displaystyle t}t is time, and {\displaystyle \sigma }\sigma ,
{\displaystyle \rho }\rho , {\displaystyle \beta }\beta are the system parameters.
Five of the terms on the right hand side are linear, while two are quadratic; a
total of seven terms. Another well-known chaotic attractor is generated by the
Rössler equations, which have only one nonlinear term out of seven. Sprott[42]
found a three-dimensional system with just five terms, that had only one nonlinear
term, which exhibits chaos for certain parameter values. Zhang and Heidel[43][44]
showed that, at least for dissipative and conservative quadratic systems, three-
dimensional quadratic systems with only three or four terms on the right-hand side
cannot exhibit chaotic behavior. The reason is, simply put, that solutions to such
systems are asymptotic to a two-dimensional surface and therefore solutions are
well behaved.

While the Poincaré–Bendixson theorem shows that a continuous dynamical system on

the Euclidean plane cannot be chaotic, two-dimensional continuous systems with non-
Euclidean geometry can exhibit chaotic behavior.[45][self-published source?]
Perhaps surprisingly, chaos may occur also in linear systems, provided they are
infinite dimensional.[46] A theory of linear chaos is being developed in a branch
of mathematical analysis known as functional analysis.

Infinite dimensional maps

The straightforward generalization of coupled discrete maps[47] is based upon
convolution integral which mediates interaction between spatially distributed maps:
{\displaystyle \psi _{n+1}({\vec {r}},t)=\int K({\vec {r}}-{\vec {r}}^{,},t)f[\psi
_{n}({\vec {r}}^{,},t)]d{\vec {r}}^{,}}{\displaystyle \psi _{n+1}({\vec
{r}},t)=\int K({\vec {r}}-{\vec {r}}^{,},t)f[\psi _{n}({\vec {r}}^{,},t)]d{\vec
{r}}^{,}},

where kernel {\displaystyle K({\vec {r}}-{\vec {r}}^{,},t)}{\displaystyle K({\vec

{r}}-{\vec {r}}^{,},t)} is propagator derived as Green function of a relevant
physical system,[48] {\displaystyle f[\psi _{n}({\vec {r}},t)]}{\displaystyle
f[\psi _{n}({\vec {r}},t)]} might be logistic map alike {\displaystyle \psi
\rightarrow G\psi [1-\tanh(\psi )]}{\displaystyle \psi \rightarrow G\psi
[1-\tanh(\psi )]} or complex map. For examples of complex maps the Julia set
{\displaystyle f[\psi ]=\psi ^{2}}{\displaystyle f[\psi ]=\psi ^{2}} or Ikeda map
{\displaystyle \psi _{n+1}=A+B\psi _{n}e^{i(|\psi _{n}|^{2}+C)}}{\displaystyle \psi
_{n+1}=A+B\psi _{n}e^{i(|\psi _{n}|^{2}+C)}} may serve. When wave propagation
problems at distance {\displaystyle L=ct}{\displaystyle L=ct} with wavelength
{\displaystyle \lambda =2\pi /k}\lambda =2\pi /k are considered the kernel
{\displaystyle K}K may have a form of Green function for Schrödinger equation:.[49]
[50]

{\displaystyle K({\vec {r}}-{\vec {r}}^{,},L)={\frac {ik\exp[ikL]}{2\pi

L}}\exp[{\frac {ik|{\vec {r}}-{\vec {r}}^{,}|^{2}}{2L}}]}{\displaystyle K({\vec
{r}}-{\vec {r}}^{,},L)={\frac {ik\exp[ikL]}{2\pi L}}\exp[{\frac {ik|{\vec {r}}-
{\vec {r}}^{,}|^{2}}{2L}}]}.

Jerk systems
In physics, jerk is the third derivative of position, with respect to time. As
such, differential equations of the form

{\displaystyle J\left({\overset {...}{x}},{\ddot {x}},{\dot

{x}},x\right)=0}J\left({\overset {...}{x}},{\ddot {x}},{\dot {x}},x\right)=0
are sometimes called jerk equations. It has been shown that a jerk equation, which
is equivalent to a system of three first order, ordinary, non-linear differential
equations, is in a certain sense the minimal setting for solutions showing chaotic
behaviour. This motivates mathematical interest in jerk systems. Systems involving
a fourth or higher derivative are called accordingly hyperjerk systems.[51]

A jerk system's behavior is described by a jerk equation, and for certain jerk
equations, simple electronic circuits can model solutions. These circuits are known
as jerk circuits.

One of the most interesting properties of jerk circuits is the possibility of

chaotic behavior. In fact, certain well-known chaotic systems, such as the Lorenz
attractor and the Rössler map, are conventionally described as a system of three
first-order differential equations that can combine into a single (although rather
complicated) jerk equation. Another example of a jerk equation with nonlinearity in
the magnitude of {\displaystyle x}x is:

{\displaystyle {\frac {\mathrm {d} ^{3}x}{\mathrm {d} t^{3}}}+A{\frac {\mathrm {d}

^{2}x}{\mathrm {d} t^{2}}}+{\frac {\mathrm {d} x}{\mathrm {d} t}}-|x|+1=0.}{\frac
{\mathrm {d} ^{3}x}{\mathrm {d} t^{3}}}+A{\frac {\mathrm {d} ^{2}x}{\mathrm {d}
t^{2}}}+{\frac {\mathrm {d} x}{\mathrm {d} t}}-|x|+1=0.
Here, A is an adjustable parameter. This equation has a chaotic solution for A=3/5
and can be implemented with the following jerk circuit; the required nonlinearity
is brought about by the two diodes:

JerkCircuit01.png
In the above circuit, all resistors are of equal value, except {\displaystyle
R_{A}=R/A=5R/3}R_{A}=R/A=5R/3, and all capacitors are of equal size. The dominant
frequency is {\displaystyle 1/2\pi RC}1/2\pi RC. The output of op amp 0 will
correspond to the x variable, the output of 1 corresponds to the first derivative
of x and the output of 2 corresponds to the second derivative.

Similar circuits only require one diode[52] or no diodes at all.[53]

See also the well-known Chua's circuit, one basis for chaotic true random number
generators.[54] The ease of construction of the circuit has made it a ubiquitous
real-world example of a chaotic system.

Spontaneous order
Under the right conditions, chaos spontaneously evolves into a lockstep pattern. In
the Kuramoto model, four conditions suffice to produce synchronization in a chaotic
system. Examples include the coupled oscillation of Christiaan Huygens' pendulums,
fireflies, neurons, the London Millennium Bridge resonance, and large arrays of
Josephson junctions.[55]

History

Barnsley fern created using the chaos game. Natural forms (ferns, clouds,
mountains, etc.) may be recreated through an iterated function system (IFS).
An early proponent of chaos theory was Henri Poincaré. In the 1880s, while studying
the three-body problem, he found that there can be orbits that are nonperiodic, and
yet not forever increasing nor approaching a fixed point.[56][57][58] In 1898,
Jacques Hadamard published an influential study of the chaotic motion of a free
particle gliding frictionlessly on a surface of constant negative curvature, called
"Hadamard's billiards".[59] Hadamard was able to show that all trajectories are
unstable, in that all particle trajectories diverge exponentially from one another,
with a positive Lyapunov exponent.

Chaos theory began in the field of ergodic theory. Later studies, also on the topic
of nonlinear differential equations, were carried out by George David Birkhoff,[60]
Andrey Nikolaevich Kolmogorov,[61][62][63] Mary Lucy Cartwright and John Edensor
Littlewood,[64] and Stephen Smale.[65] Except for Smale, these studies were all
directly inspired by physics: the three-body problem in the case of Birkhoff,
turbulence and astronomical problems in the case of Kolmogorov, and radio
engineering in the case of Cartwright and Littlewood.[citation needed] Although
chaotic planetary motion had not been observed, experimentalists had encountered
turbulence in fluid motion and nonperiodic oscillation in radio circuits without
the benefit of a theory to explain what they were seeing.

Despite initial insights in the first half of the twentieth century, chaos theory
became formalized as such only after mid-century, when it first became evident to
some scientists that linear theory, the prevailing system theory at that time,
simply could not explain the observed behavior of certain experiments like that of
the logistic map. What had been attributed to measure imprecision and simple
"noise" was considered by chaos theorists as a full component of the studied
systems.

The main catalyst for the development of chaos theory was the electronic computer.
Much of the mathematics of chaos theory involves the repeated iteration of simple
mathematical formulas, which would be impractical to do by hand. Electronic
computers made these repeated calculations practical, while figures and images made
it possible to visualize these systems. As a graduate student in Chihiro Hayashi's
laboratory at Kyoto University, Yoshisuke Ueda was experimenting with analog
computers and noticed, on November 27, 1961, what he called "randomly transitional
phenomena". Yet his advisor did not agree with his conclusions at the time, and did
not allow him to report his findings until 1970.[66][67]
Turbulence in the tip vortex from an airplane wing. Studies of the critical point
beyond which a system creates turbulence were important for chaos theory, analyzed
for example by the Soviet physicist Lev Landau, who developed the Landau-Hopf
theory of turbulence. David Ruelle and Floris Takens later predicted, against
Landau, that fluid turbulence could develop through a strange attractor, a main
concept of chaos theory.
Edward Lorenz was an early pioneer of the theory. His interest in chaos came about
accidentally through his work on weather prediction in 1961.[12] Lorenz was using a
simple digital computer, a Royal McBee LGP-30, to run his weather simulation. He
wanted to see a sequence of data again, and to save time he started the simulation
in the middle of its course. He did this by entering a printout of the data that
corresponded to conditions in the middle of the original simulation. To his
surprise, the weather the machine began to predict was completely different from
the previous calculation. Lorenz tracked this down to the computer printout. The
computer worked with 6-digit precision, but the printout rounded variables off to a
3-digit number, so a value like 0.506127 printed as 0.506. This difference is tiny,
and the consensus at the time would have been that it should have no practical
effect. However, Lorenz discovered that small changes in initial conditions
produced large changes in long-term outcome.[68] Lorenz's discovery, which gave its
name to Lorenz attractors, showed that even detailed atmospheric modelling cannot,
in general, make precise long-term weather predictions.

In 1963, Benoit Mandelbrot found recurring patterns at every scale in data on

cotton prices.[69] Beforehand he had studied information theory and concluded noise
was patterned like a Cantor set: on any scale the proportion of noise-containing
periods to error-free periods was a constant – thus errors were inevitable and must
be planned for by incorporating redundancy.[70] Mandelbrot described both the "Noah
effect" (in which sudden discontinuous changes can occur) and the "Joseph effect"
(in which persistence of a value can occur for a while, yet suddenly change
afterwards).[71][72] This challenged the idea that changes in price were normally
distributed. In 1967, he published "How long is the coast of Britain? Statistical
self-similarity and fractional dimension", showing that a coastline's length varies
with the scale of the measuring instrument, resembles itself at all scales, and is
infinite in length for an infinitesimally small measuring device.[73] Arguing that
a ball of twine appears as a point when viewed from far away (0-dimensional), a
ball when viewed from fairly near (3-dimensional), or a curved strand (1-
dimensional), he argued that the dimensions of an object are relative to the
observer and may be fractional. An object whose irregularity is constant over
different scales ("self-similarity") is a fractal (examples include the Menger
sponge, the Sierpiński gasket, and the Koch curve or snowflake, which is infinitely
long yet encloses a finite space and has a fractal dimension of circa 1.2619). In
1982, Mandelbrot published The Fractal Geometry of Nature, which became a classic
of chaos theory.[74] Biological systems such as the branching of the circulatory
and bronchial systems proved to fit a fractal model.[75]

In December 1977, the New York Academy of Sciences organized the first symposium on
chaos, attended by David Ruelle, Robert May, James A. Yorke (coiner of the term
"chaos" as used in mathematics), Robert Shaw, and the meteorologist Edward Lorenz.
The following year Pierre Coullet and Charles Tresser published "Itérations
d'endomorphismes et groupe de renormalisation", and Mitchell Feigenbaum's article
"Quantitative Universality for a Class of Nonlinear Transformations" finally
appeared in a journal, after 3 years of referee rejections.[39][76] Thus Feigenbaum
(1975) and Coullet & Tresser (1978) discovered the universality in chaos,
permitting the application of chaos theory to many different phenomena.

In 1979, Albert J. Libchaber, during a symposium organized in Aspen by Pierre

Hohenberg, presented his experimental observation of the bifurcation cascade that
leads to chaos and turbulence in Rayleigh–Bénard convection systems. He was awarded
the Wolf Prize in Physics in 1986 along with Mitchell J. Feigenbaum for their
inspiring achievements.[77]

In 1986, the New York Academy of Sciences co-organized with the National Institute
of Mental Health and the Office of Naval Research the first important conference on
chaos in biology and medicine. There, Bernardo Huberman presented a mathematical
model of the eye tracking disorder among schizophrenics.[78] This led to a renewal
of physiology in the 1980s through the application of chaos theory, for example, in
the study of pathological cardiac cycles.

In 1987, Per Bak, Chao Tang and Kurt Wiesenfeld published a paper in Physical
Review Letters[79] describing for the first time self-organized criticality (SOC),
considered one of the mechanisms by which complexity arises in nature.

Alongside largely lab-based approaches such as the Bak–Tang–Wiesenfeld sandpile,

many other investigations have focused on large-scale natural or social systems
that are known (or suspected) to display scale-invariant behavior. Although these
approaches were not always welcomed (at least initially) by specialists in the
subjects examined, SOC has nevertheless become established as a strong candidate
for explaining a number of natural phenomena, including earthquakes, (which, long
before SOC was discovered, were known as a source of scale-invariant behavior such
as the Gutenberg–Richter law describing the statistical distribution of earthquake
sizes, and the Omori law[80] describing the frequency of aftershocks), solar
flares, fluctuations in economic systems such as financial markets (references to
SOC are common in econophysics), landscape formation, forest fires, landslides,
epidemics, and biological evolution (where SOC has been invoked, for example, as
the dynamical mechanism behind the theory of "punctuated equilibria" put forward by
Niles Eldredge and Stephen Jay Gould). Given the implications of a scale-free
distribution of event sizes, some researchers have suggested that another
phenomenon that should be considered an example of SOC is the occurrence of wars.
These investigations of SOC have included both attempts at modelling (either
developing new models or adapting existing ones to the specifics of a given natural
system), and extensive data analysis to determine the existence and/or
characteristics of natural scaling laws.

In the same year, James Gleick published Chaos: Making a New Science, which became
a best-seller and introduced the general principles of chaos theory as well as its
history to the broad public, though his history under-emphasized important Soviet
contributions.[citation needed][81] Initially the domain of a few, isolated
individuals, chaos theory progressively emerged as a transdisciplinary and
institutional discipline, mainly under the name of nonlinear systems analysis.
Alluding to Thomas Kuhn's concept of a paradigm shift exposed in The Structure of
Scientific Revolutions (1962), many "chaologists" (as some described themselves)
claimed that this new theory was an example of such a shift, a thesis upheld by
Gleick.

The availability of cheaper, more powerful computers broadens the applicability of

chaos theory. Currently, chaos theory remains an active area of research,[82]
involving many different disciplines such as mathematics, topology, physics,[83]
social systems,[84] population modeling, biology, meteorology, astrophysics,
information theory, computational neuroscience, pandemic crisis management,[16][17]
etc.

Applications

A conus textile shell, similar in appearance to Rule 30, a cellular automaton with
chaotic behaviour.[85]
Although chaos theory was born from observing weather patterns, it has become
applicable to a variety of other situations. Some areas benefiting from chaos
theory today are geology, mathematics, biology, computer science, economics,[86]
[87][88] engineering,[89][90] finance,[91][92] algorithmic trading,[93][94][95]
meteorology, philosophy, anthropology,[15] physics,[96][97][98] politics,[99][100]
population dynamics,[101] psychology,[14] and robotics. A few categories are listed
below with examples, but this is by no means a comprehensive list as new
applications are appearing.

Cryptography
Chaos theory has been used for many years in cryptography. In the past few decades,
chaos and nonlinear dynamics have been used in the design of hundreds of
cryptographic primitives. These algorithms include image encryption algorithms,
hash functions, secure pseudo-random number generators, stream ciphers,
watermarking and steganography.[102] The majority of these algorithms are based on
uni-modal chaotic maps and a big portion of these algorithms use the control
parameters and the initial condition of the chaotic maps as their keys.[103] From a
wider perspective, without loss of generality, the similarities between the chaotic
maps and the cryptographic systems is the main motivation for the design of chaos
based cryptographic algorithms.[102] One type of encryption, secret key or
symmetric key, relies on diffusion and confusion, which is modeled well by chaos
theory.[104] Another type of computing, DNA computing, when paired with chaos
theory, offers a way to encrypt images and other information.[105] Many of the DNA-
Chaos cryptographic algorithms are proven to be either not secure, or the technique
applied is suggested to be not efficient.[106][107][108]

Robotics
Robotics is another area that has recently benefited from chaos theory. Instead of
robots acting in a trial-and-error type of refinement to interact with their
environment, chaos theory has been used to build a predictive model.[109] Chaotic
dynamics have been exhibited by passive walking biped robots.[110]

Biology
For over a hundred years, biologists have been keeping track of populations of
different species with population models. Most models are continuous, but recently
scientists have been able to implement chaotic models in certain populations.[111]
For example, a study on models of Canadian lynx showed there was chaotic behavior
in the population growth.[112] Chaos can also be found in ecological systems, such
as hydrology. While a chaotic model for hydrology has its shortcomings, there is
still much to learn from looking at the data through the lens of chaos theory.[113]
Another biological application is found in cardiotocography. Fetal surveillance is
a delicate balance of obtaining accurate information while being as noninvasive as
possible. Better models of warning signs of fetal hypoxia can be obtained through
chaotic modeling.[114]

Economics
It is possible that economic models can also be improved through an application of
chaos theory, but predicting the health of an economic system and what factors
influence it most is an extremely complex task.[115] Economic and financial systems
are fundamentally different from those in the classical natural sciences since the
former are inherently stochastic in nature, as they result from the interactions of
people, and thus pure deterministic models are unlikely to provide accurate
representations of the data. The empirical literature that tests for chaos in
economics and finance presents very mixed results, in part due to confusion between
specific tests for chaos and more general tests for non-linear relationships.[116]

Chaos could be found in economics by the means of recurrence quantification

analysis. In fact, Orlando et al.[117] by the means of the so-called recurrence
quantification correlation index were able detect hidden changes in time series.
Then, the same technique was employed to detect transitions from laminar (i.e.
regular) to turbulent (i.e. chaotic) phases as well as differences between
macroeconomic variables and highlight hidden features of economic dynamics.[118]
Finally, chaos could help in modeling how economy operate as well as in embedding
shocks due to external events such as COVID-19.[119] For an updated account on the
tools and the results obtained by empirically calibrating and testing deterministic
chaotic models (e.g. Kaldor-Kalecki,[120] Goodwin,[121] Harrod [122]), see Orlando
et al.[123]

Other areas
In chemistry, predicting gas solubility is essential to manufacturing polymers, but
models using particle swarm optimization (PSO) tend to converge to the wrong
points. An improved version of PSO has been created by introducing chaos, which
keeps the simulations from getting stuck.[124] In celestial mechanics, especially
when observing asteroids, applying chaos theory leads to better predictions about
when these objects will approach Earth and other planets.[125] Four of the five
moons of Pluto rotate chaotically. In quantum physics and electrical engineering,
the study of large arrays of Josephson junctions benefitted greatly from chaos
theory.[126] Closer to home, coal mines have always been dangerous places where
frequent natural gas leaks cause many deaths. Until recently, there was no reliable
way to predict when they would occur. But these gas leaks have chaotic tendencies
that, when properly modeled, can be predicted fairly accurately.[127]

Chaos theory can be applied outside of the natural sciences, but historically
nearly all such studies have suffered from lack of reproducibility; poor external
validity; and/or inattention to cross-validation, resulting in poor predictive
accuracy (if out-of-sample prediction has even been attempted). Glass[128] and
Mandell and Selz[129] have found that no EEG study has as yet indicated the
presence of strange attractors or other signs of chaotic behavior.

Researchers have continued to apply chaos theory to psychology. For example, in

modeling group behavior in which heterogeneous members may behave as if sharing to
different degrees what in Wilfred Bion's theory is a basic assumption, researchers
have found that the group dynamic is the result of the individual dynamics of the
members: each individual reproduces the group dynamics in a different scale, and
the chaotic behavior of the group is reflected in each member.[130]

Redington and Reidbord (1992) attempted to demonstrate that the human heart could
display chaotic traits. They monitored the changes in between-heartbeat intervals
for a single psychotherapy patient as she moved through periods of varying
emotional intensity during a therapy session. Results were admittedly inconclusive.
Not only were there ambiguities in the various plots the authors produced to
purportedly show evidence of chaotic dynamics (spectral analysis, phase trajectory,
and autocorrelation plots), but also when they attempted to compute a Lyapunov
exponent as more definitive confirmation of chaotic behavior, the authors found
they could not reliably do so.[131]

In their 1995 paper, Metcalf and Allen[132] maintained that they uncovered in
animal behavior a pattern of period doubling leading to chaos. The authors examined
a well-known response called schedule-induced polydipsia, by which an animal
deprived of food for certain lengths of time will drink unusual amounts of water
when the food is at last presented. The control parameter (r) operating here was
the length of the interval between feedings, once resumed. The authors were careful
to test a large number of animals and to include many replications, and they
designed their experiment so as to rule out the likelihood that changes in response
patterns were caused by different starting places for r.

Time series and first delay plots provide the best support for the claims made,
showing a fairly clear march from periodicity to irregularity as the feeding times
were increased. The various phase trajectory plots and spectral analyses, on the
other hand, do not match up well enough with the other graphs or with the overall
theory to lead inexorably to a chaotic diagnosis. For example, the phase
trajectories do not show a definite progression towards greater and greater
complexity (and away from periodicity); the process seems quite muddied. Also,
where Metcalf and Allen saw periods of two and six in their spectral plots, there
is room for alternative interpretations. All of this ambiguity necessitate some
serpentine, post-hoc explanation to show that results fit a chaotic model.

By adapting a model of career counseling to include a chaotic interpretation of the

relationship between employees and the job market, Amundson and Bright found that
better suggestions can be made to people struggling with career decisions.[133]
Modern organizations are increasingly seen as open complex adaptive systems with
fundamental natural nonlinear structures, subject to internal and external forces
that may contribute chaos. For instance, team building and group development is
increasingly being researched as an inherently unpredictable system, as the
uncertainty of different individuals meeting for the first time makes the
trajectory of the team unknowable.[134]

Some say the chaos metaphor—used in verbal theories—grounded on mathematical models

and psychological aspects of human behavior provides helpful insights to describing
the complexity of small work groups, that go beyond the metaphor itself.[135]

The red cars and blue cars take turns to move; the red ones only move upwards, and
the blue ones move rightwards. Every time, all the cars of the same colour try to
move one step if there is no car in front of it. Here, the model has self-organized
in a somewhat geometric pattern where there are some traffic jams and some areas
where cars can move at top speed.
Traffic forecasting may benefit from applications of chaos theory. Better
predictions of when traffic will occur would allow measures to be taken to disperse
it before it would have occurred. Combining chaos theory principles with a few
other methods has led to a more accurate short-term prediction model (see the plot
of the BML traffic model at right).[136]

Chaos theory has been applied to environmental water cycle data (aka hydrological
data), such as rainfall and streamflow.[137] These studies have yielded
controversial results, because the methods for detecting a chaotic signature are
often relatively subjective. Early studies tended to "succeed" in finding chaos,
whereas subsequent studies and meta-analyses called those studies into question and
provided explanations for why these datasets are not likely to have low-dimension
chaotic dynamics.[138]

See also
diagram Systems science portal
icon Mathematics portal
Examples of chaotic systems
Advected contours
Arnold's cat map
Bouncing ball dynamics
Chua's circuit
Cliodynamics
Coupled map lattice
Double pendulum
Duffing equation
Dynamical billiards
Economic bubble
Gaspard-Rice system
Hénon map
Horseshoe map
List of chaotic maps
Rössler attractor
Standard map
Swinging Atwood's machine
Tilt A Whirl
Other related topics
Amplitude death
Anosov diffeomorphism
Catastrophe theory
Causality
Chaos theory in organizational development
Chaos machine
Chaotic mixing
Chaotic scattering
Control of chaos
Determinism
Edge of chaos
Emergence
Mandelbrot set
Kolmogorov–Arnold–Moser theorem
Ill-conditioning
Ill-posedness
Nonlinear system
Patterns in nature
Predictability
Quantum chaos
Santa Fe Institute
Synchronization of chaos
Unintended consequence
People
Ralph Abraham
Michael Berry
Leon O. Chua
Ivar Ekeland
Doyne Farmer
Martin Gutzwiller
Brosl Hasslacher
Michel Hénon
Aleksandr Lyapunov
Norman Packard
Otto Rössler
David Ruelle
Oleksandr Mikolaiovich Sharkovsky
Robert Shaw
Floris Takens
James A. Yorke
George M. Zaslavsky
References