International Journal of Advance Research, IJOAR .

org
ISSN xxxx-xxxx

IJOAR 2013
http://www.ijoar.org
Fractal: A set which is larger than the universe
Soumya Prakash Sahu, Indian Institute of Science Education & Research,
Bhopal
Email id- [email protected]
Keyword list
Fractal, Fractal dimension, Chaos theory, Dynamical system, Attractor, Space-
filling curve, the Mandelbrot set and Julia set
Abstract
Fractal is a set, which geometric pattern is self-similar at different scales. It has
a fractal dimension, which strictly exceeds its topological dimension. It can be
a non-integer. We calculate fractal dimension for only exact self-similar
fractals. We show a strong link between chaos theory and fractals in an
informal manner and study about space-filling curves, which are special type of
fractals having fractal dimension two. Fractals need not only to be self- similar
in nature. We explain about different types of fractals depending upon their
geometric patterns. Finally we discuss about the Mandelbrot set, one of the
most complex structure in mathematics and the Julia set. The Julia set can be
connected or disconnected. The connectedness of the Julia set depends on the
structure of the Mandelbrot set.












International Journal of Advance Research, IJOAR .org
ISSN xxxx-xxxx

IJOAR 2013
http://www.ijoar.org
1. Introduction
Clouds are not spheres, mountains are not cones, coastlines are not
circles, and bark is not smooth, nor does lightning travel in a straight line.
Benot Mandelbrot
It is hard to fathom how extraordinary beautiful patterns like cloud
formations, the veins of leaves and the branching of trees are so ubiquitous in
nature. Surprisingly, these naturally occurring self-similar intricate patterns
have a connection to mathematics. Benot Mandelbrot, professor at Yale
University, first studied these self-similar patterns in nature in a graphical and
structured manner in 1970s. He called these patterns fractals. hough before
him man other mathematicians lie eorg antor, aston ulia, enri
oincar e and elge on Koch studied these patterns, their analysis was
limited to what they could depict in manual drawings.
To understand fractals, let us first ask ourselves the following question:
How long is the coast of India? To measure this, take a fixed finite length x and
move along the coastline in steps of length x until we reach or cross the end of
the coastline. Multiplying the number of steps we took before we reached the
end with x, we obtain a very crude approximation L(x) for the length of the
coastline. It is true that by reducing x, we could possibly improve the
approximation, but no matter how hard we try to better our estimate, the
infinite irregularities in the coastline would render this endeavor rather futile.
That means the length of the coastline is infinite having finite area [4]! In other
words, the coastline is a fractal.
What is a fractal? A fractal is a spatial object whose geometric
characteristics include irregularity, scale-independence and self-similarity [4].
Self-similarity means the fractal repeats itself at different scales (dilation?) [4].
Generally people have the idea that a continuous smooth curve has a unique
tangent at every point on it except at the corners but most fractals are
counterintuitive as these are continuous and nowhere differentiable curves due
to their irregularity. Therefore, fractals are pathological objects, that is, they are
very counterintuitive or badly behaved [4].

2. Fractal dimension
Formally, a fractal is a mathematical set having a fractal dimension,
which strictly exceeds its topological dimension [1]. The fractal dimension is a
number associated with every fractal that measures the roughness of the fractal
[4]. Informally, the number of independent directions occupied by an object in
space is the dimension of that object. For example, a line is 1-dimensional, a
square is 2-dimensional and a cube is 3-dimensional. The motion of any body
in phase-space requires three position coordinates and three momentum
International Journal of Advance Research, IJOAR .org
ISSN xxxx-xxxx

IJOAR 2013
http://www.ijoar.org
coordinates. Our finger can be bent by many independent angles and it requires
10 coordinates to describe its state perfectly [4]. In the pure mathematical
subject of topology, many objects can be continuously deformed to one another
but will continue to have the same dimension [3]. This type of dimension is
called topological dimension [4]. For a better understanding of the fractal
dimension, we need to know a little bit more than just its topological
properties. So the fractal dimension is defined in a new way taking distance
(scaling properties) or metric properties into consideration.
Naively, a metric space is a set where distance between elements of the
set is defined [2]. Any subset of the metric space (H(X),h) is a fractal. Here h is
the Hausdorff metric, (X,d) is a complete metric space and points of H(X)
space are the compact subsets of X, a non-empty set [1]. (There is a difference
between any non-empty compact subset of X and any subset of H(X).) [1].
Hausdorff dimension is a non-negative real number associated with every
metric space. Felix Hausdorff first introduced it in 1918.
Fractals can have non-integer dimensions. Fractal dimension can be
calculated by the formula d = log(N)/log(n) [4]. Where linear size of that object
can be reduced by 1/n

in each spatial direction and N is the increase of the
measure (length, area, volume) by n^d times from its original value and d is the
fractal dimension. For example, the Koch curve has fractal dimension d =
log(4)/log(3)= 1.26, the Cantor set has d = log(3)/log(2) = 0.6309. But this
definition of dimension is restricted only to self-similar fractals i.e., it must
coincide with small piece of itself when suitably magnified.
3. Chaos theory and fractals
The study of fractals becomes more interesting when we study chaos
theory. Chaos theory is the study of unstable aperiodic behavior on nonlinear
dynamical systems like weather systems [2]. Naively, a dynamical system is
any system that changes dynamically with time. Nonlinear (deterministic
chaos) dynamical system means that the output of the system does not depend
directly on the input and any changes occurring at inputs does not affect the
variables or the output at the same time [2]. Chaos deals with dissipative
(energy loss that happens as a function of time) nonlinear dynamical systems
and can be represented by nonlinear differential equations. Edward Lorenz, a
meteorologist, first discovered chaos theory when he described the butterfly
effect in 1961. Butterfly effect describes about the sensitive dependence of a
system on initial conditions, which is one of the cause for the chaotic motion.
An attractor is a set of stable conditions of a dynamical system [2]. An
attractor can be chaotic (strange) or non-chaotic. A fractal can describe the
geometric nature of an attractor. Chaotic attractors are the attractors with a
fractal structure. This is how chaos theory and fractals are related.
4. Types of fractals
Fractals need not always be self-similar since they can also be self-
affine in nature. Self-affine means that the ratio of self-similarity in different
International Journal of Advance Research, IJOAR .org
ISSN xxxx-xxxx

IJOAR 2013
http://www.ijoar.org
directions is different. Self-similarity can also be of several types such as exact
self-similarity, quasi-self-similarity and statistical self-similarity. Exact self-
similarity means that the fractal appears exactly identical, quasi self-similarity
means that the fractal appears approximately identical and statistical self-
similarity means that the fractal preserves its statistical similarity at different
scales. The coast of India has statistical self-similarity and Koch Snowflake has
exact self-similarity.
Another type of fractal is the space-filling curve. Space-filling curve is
a continuous function with do-main [0, 1] and range is a 2-dimensional unit
square (or more generally an N-dimensional hypercube). Intuitively, it is a
continuous curve in 2 or 3 (or higher) dimensions and can be thought of as the
path of a continuously moving point. For example the Peano curve which
covers all the points in a unit square and snowflake sweep, which fills the Koch
snowflake. Both the curves have fractal dimension two. There are other types
of fractals like Base-Motif fractals, Julia set, IFS (Iterated Function System)
fractals, plasma fractals, fractal canopies, paper-folding fractals, shape-
replacement fractals etc. Several types like iteration, chaos game, fractal
transformation; l-system, quaternion and strange attractors can construct these
types of fractals.
5. Mandelbrot set and Julia set
Lets talk about one of the most complex structures of mathematics, the
Mandelbrot set. It is a quasi- self-similarity type of fractal. It can be
constructed by a formula iteration. The formula for the iteration is F (z) =
z^2+c. The iteration is done in a complex plane, where the input is z=0+i0 (the
critical point of F(z)) and c is any complex number [5]. If the output tends to
zero (which means F(z) at z=0+0i is bounded), the point c is contained within
the set (the boundary of Mandelbrot set) and if the output tends to infinity, then
c is outside the set [5].
The Julia set is the boundary of the filled-in Julia set. The filled-in Julia
set is formed by a collection of points (z) for which F(z) do not tend to infinity
after infinite iterations for an arbitrary c (the same c as in the Mandelbrot Set).
The Julia set can be connected or disconnected [5]. Although both the sets are
different, the interconnected nature of the Julia set depends on the Mandelbrot
set in the following manner. The Mandelbrot set is the set of c, for which the
Julia set is connected.
6. Conclusion
The theory of fractals can be applied to several fields like astronomy,
biology, chemistry, art, music, weather, computer graphics, animation and data
compression. It is now also used for Euclidean tessellation.

International Journal of Advance Research, IJOAR .org
ISSN xxxx-xxxx

IJOAR 2013
http://www.ijoar.org
Acknowledgements
I would like to thank my fellow friend Arghyadeep Dash and my
advisor Dr. Kashyap Rajeevsarathy for helping me to write this article.
References
[1] Barnsley, Michael F., and Hawley Rising. Fractals Everywhere. Boston:
Academic Press Professional, 1993.
[2] Ahmed M. Saeed, Shimaa A. Abd El Moat, Victor I. Fahmy. Chaos and
Fractals.
[3] Arghyadeep Dash, Kunal Sharma, Soumya Prakash Sahu guided by Dr.
Kashyap Rajeevsarathy. Exploring Surfaces. Geodesics, The annual newsletter
of Indian Institute of Science Education & Research, Bhopal, ISSUE I,
February 2013.

[4] Arthur C. Clarke, Benot Mandelbrot (Emeritus), Dr. David Pennock, Dr.
Gary Flake, Ian Stewart (Emeritus),Michael Barnsley, Nigel Lesmoir-Gordon,
Will Rood. The Colours of Infinity. Springer.
[5] Theodore W. Gamelin. Complex Analysis. Springer