Fractals and Chaos Theory

Clouds are not spheres, coastlines are not straight lines, and tomorrow's weather cannot be predicted a month ahead no matter how powerful the computer. Fractals and chaos theory are the branches of mathematics that explain why. They show that very simple rules, repeated again and again, can produce endless complexity and genuine unpredictability.

What Is a Fractal?

A fractal is a shape that shows detail at every scale. Zoom in on part of it and you see structure similar to the whole, a property called self-similarity. Ferns, broccoli, river networks, lightning and blood vessels all have fractal-like structure.

The word fractal was coined in 1975 by Benoit Mandelbrot, a Polish-born French-American mathematician working at IBM, from the Latin fractus, meaning "broken". His 1967 paper "How Long Is the Coast of Britain?" pointed out that a coastline's measured length keeps increasing as your ruler gets shorter, because you capture more and more wiggles.

The Koch Snowflake

In 1904 the Swedish mathematician Helge von Koch described a curve built by a simple rule:

1. Start with an equilateral triangle.
2. Divide each side into three equal parts.
3. Replace the middle part with two sides of a smaller equilateral triangle pointing outward.
4. Repeat forever.

Worked Example 1: Perimeter and area of the Koch snowflake

Problem: Start with an equilateral triangle of side 1 and area \( A_0 \). Find the perimeter after \( n \) steps and the final area.

Step 1 (perimeter). Each step replaces every segment with 4 segments, each \( \tfrac{1}{3} \) as long, so the perimeter is multiplied by \( \tfrac{4}{3} \):

$$ P_n = 3\left(\frac{4}{3}\right)^n $$

Since \( \tfrac{4}{3} > 1 \), \( P_n \to \infty \). After 10 steps, \( P_{10} = 3 \times (4/3)^{10} \approx 3 \times 17.76 \approx 53.3 \).

Step 2 (new triangles). At step 1 we add 3 triangles, each with side \( \tfrac{1}{3} \) and so area \( \tfrac{1}{9}A_0 \). At each later step the number of sides multiplies by 4 and each new triangle's area is \( \tfrac{1}{9} \) of the previous ones. So the added areas form a geometric series:

$$ A = A_0 + \frac{3}{9}A_0 + \frac{12}{81}A_0 + \frac{48}{729}A_0 + \cdots = A_0 + \frac{1}{3}A_0\left(1 + \frac{4}{9} + \left(\frac{4}{9}\right)^2 + \cdots\right) $$

Step 3. Sum the geometric series with ratio \( \tfrac{4}{9} \):

$$ 1 + \frac{4}{9} + \left(\frac{4}{9}\right)^2 + \cdots = \frac{1}{1 - 4/9} = \frac{9}{5} $$

Step 4. Therefore

$$ A = A_0 + \frac{1}{3} \cdot \frac{9}{5} A_0 = A_0 + \frac{3}{5}A_0 = \frac{8}{5}A_0 $$

The snowflake has an infinite perimeter enclosing a finite area, only 1.6 times the original triangle. That paradox is typical of fractals.

Fractal Dimension

A line has dimension 1, a square 2, a cube 3. If you scale a shape by a factor of \( \tfrac{1}{s} \) and need \( N \) copies to rebuild it, the similarity dimension is

$$ D = \frac{\log N}{\log s} $$

Check: a square split into pieces of side \( \tfrac{1}{3} \) needs \( N = 9 \) copies, so \( D = \log 9 / \log 3 = 2 \), as expected.

Worked Example 2: Dimension of the Koch curve and Sierpinski triangle

Koch curve. Each piece is scaled by \( \tfrac{1}{3} \) and there are 4 pieces:

$$ D = \frac{\log 4}{\log 3} = \frac{0.60206}{0.47712} \approx 1.262 $$

Sierpinski triangle. Each piece is scaled by \( \tfrac{1}{2} \) and there are 3 pieces:

$$ D = \frac{\log 3}{\log 2} = \frac{0.47712}{0.30103} \approx 1.585 $$

These non-whole dimensions measure how "rough" or "space-filling" a shape is. The Koch curve is more than a line but less than a surface.

What Is Chaos?

Chaos describes systems that follow exact, deterministic rules yet behave unpredictably because tiny differences in the starting conditions grow rapidly. This is sensitive dependence on initial conditions.

In 1961 the American meteorologist Edward Lorenz, at MIT, re-ran a weather simulation by typing in a number rounded to three decimal places (0.506 instead of 0.506127). Within a few simulated weeks the forecast was completely different. His 1972 talk title, asking whether a butterfly flapping its wings in Brazil could set off a tornado in Texas, gave us the phrase "butterfly effect".

The Logistic Map

The simplest chaotic system is the logistic map, popularised by the biologist Robert May in 1976 as a model of population:

$$ x_{n+1} = r\,x_n(1 - x_n) $$

where \( x_n \) is the population as a fraction of the maximum, and \( r \) is a growth parameter.

Worked Example 3: Two nearly identical starts

Problem: With \( r = 4 \), compare starting values \( x_0 = 0.2 \) and \( y_0 = 0.2001 \).

Step 1. First iteration:

$$ x_1 = 4(0.2)(0.8) = 0.64, \qquad y_1 = 4(0.2001)(0.7999) \approx 0.6402 $$

Step 2. Second iteration:

$$ x_2 = 4(0.64)(0.36) = 0.9216, \qquad y_2 \approx 4(0.6402)(0.3598) \approx 0.9213 $$

Step 3. Third iteration:

$$ x_3 = 4(0.9216)(0.0784) \approx 0.2890, \qquad y_3 \approx 0.2899 $$

Step 4. Continuing on a computer, by \( n = 10 \) we get \( x_{10} \approx 0.148 \) but \( y_{10} \approx 0.249 \), and by \( n = 15 \) the values are about 0.004 and 0.723, completely unrelated. A starting difference of 0.0001 roughly doubles at every step on average, so it soon swamps everything. This is why long-range forecasts are impossible even with perfect equations.

For smaller \( r \) the behaviour is tame: with \( r = 2.5 \) the population settles to \( 1 - \tfrac{1}{r} = 0.6 \). As \( r \) increases past 3, the system oscillates between 2 values, then 4, then 8, a "period-doubling cascade" that leads to chaos at about \( r \approx 3.57 \).

The Mandelbrot Set

Fractals and chaos meet in the Mandelbrot set: the complex numbers \( c \) for which iterating \( z_{n+1} = z_n^2 + c \) from \( z_0 = 0 \) stays bounded. For \( c = -1 \) the sequence is \( 0, -1, 0, -1, \ldots \), which stays bounded, so \( -1 \) is in the set. For \( c = 1 \) we get \( 0, 1, 2, 5, 26, \ldots \), which escapes to infinity, so 1 is not. The boundary of the set is infinitely intricate.

Did You Know?

• The ratio between successive period-doubling points of the logistic map approaches the Feigenbaum constant \( \approx 4.669 \), the same for a huge range of systems.
• Fractal antennas in some mobile phones pack a long, wiggly conductor into a small space to receive many frequencies.
• Film studios use fractal algorithms to generate realistic mountains, coastlines and clouds.

Key Takeaways

• Fractals are self-similar shapes with detail at every scale; the term was coined by Benoit Mandelbrot in 1975.
• The Koch snowflake has infinite perimeter but finite area, \( \tfrac{8}{5} \) of the original triangle.
• Fractal dimension \( D = \log N / \log s \) can be a non-integer, such as 1.262 for the Koch curve.
• Chaos means deterministic rules with extreme sensitivity to initial conditions, the "butterfly effect" found by Edward Lorenz.
• The logistic map \( x_{n+1} = r x_n (1 - x_n) \) shows how a one-line formula can produce chaos.

Need step-by-step help with sequences, series or iterative methods? Amathify's experts and PhD scholars can solve your problems and assignments, message us on WhatsApp or use the Ask a Question form.