On a theorem for determining what is not in the filled-in Julia set of a function

73 Views Asked by At

Let $f(z) = z^2 + c$ for some $c\in\Bbb C$, and let $R=\dfrac{1+\sqrt{1+4\lvert c\rvert}}2$. If for some $a\in\Bbb C$ and $n>0$ we have that $\lvert f^n(a)\rvert>R$, then $f^n(a)\to\infty$ as $n\to\infty$ — i.e., $a$ is in the basin of attraction to infinity, so it is not in the closed Julia set.

There is a lot going on there. What I would like to know is whence this value of $R$ came and why the threshold of attraction to infinity can be simply confined to some circle.

I would try to do more research on my own, but as it was presented to me this theorem is nameless.

1

There are 1 best solutions below

1
On BEST ANSWER

Christoph Stroh in his thesis "Julia Sets of Complex Polynomials and Their Implementation on the Computer" proves that the Julia set of $f$ is contained in the disk of radius $R$ centered at the origin. See section 3.1, especially example 3.8.

The expression for $R$ comes from the largest fixed point of $f$: $$ z^2+c =z \iff z = \frac{1 - \sqrt{1 - 4 c}}{2} $$ This fixed point certainly belongs to the Julia set. Actually, this point is on the border of the convex hull of the Julia set.

The value given for $R$ allows us to prove that $|f(z)| > R$ if $|z|>R$: $$ |f (z)| = |z^2 +c| \ge |z^2|−|c| = |z|^2 −|c| > |z| > R $$ because $|z|^2 −|c| > |z|$ if $|z| > R$.

But this is not enough to prove that $|f^n(a)| \to \infty$. More later.