The recent question Area of the boundary of the Mandelbrot set ? prompted me to ask this question.

There has been some work on estimates for the area of the Mandelbrot set, e.g., a paper by John H. Ewing and Glenn Schober in Numerische Mathematik. Is there similar work for the estimating the area of quadratic filled Julia sets as a function of the parameter $c$? Perhaps the material in the book Computability of Julia Sets implies some estimates but I don't have the book handy and from what I recall of skimming it, there was none.

  • $\begingroup$ See my answer to that question - it points to work of Buff and Cheritat for the 'latest' on that question. There is also lots of earlier work on estimating the Hausdorff dimension (when it is <2 ). $\endgroup$ – Jacques Carette Sep 9 '10 at 0:39
  • $\begingroup$ I meant the area of the filled Julia sets $K$, not the Julia sets $J$. $\endgroup$ – lhf Sep 9 '10 at 2:25

This paper contains some information about the area of filled Julia sets, though not a formula:

Yang, Guoxiao, Some geometric properties of Julia sets and filled-in Julia sets of polynomials. Complex Var. Theory Appl. 47 (2002), no. 5, 383–391. MR1906990 (2003c:37067)

There is also this more promising paper by the same author, but it is in Chinese and I can't get a copy anyway:

Yang, Guo Xiao, The area and diameter of filled-in Julia sets and Mandelbrot sets. Acta Math. Sinica 38 (1995), no. 5, 607–613. MR1372560 (96m:30040)

If someone knows these papers, I'd be grateful for any insights.

Problem A-1 in Milnor's Dynamics in one complex variable contains a formula for the area expressed as a series based on Gronwall's area theorem: $$ \pi (1 - |a_2|^2 - 3|a_4|^2 - 5|a_6|^2 - \cdots) $$ The series is said to converge slowly. The coefficients of the series can be easily computed recursively though by solving $$ \psi(w^2) = \psi(w)^2+c $$ for $$ \def\F#1{\frac{a_{#1}}{w^{#1}}} \psi(w) = w(1 + \F2 + \F4 + \F6 + \cdots) $$

| cite | improve this answer | |
  • $\begingroup$ what mean the values $a_k$ and $\psi(w^2)=\psi(w)^2+c$? $\endgroup$ – user128924 Sep 16 '18 at 10:19
  • $\begingroup$ The functional equation $\psi(w^2)=\psi(w)^2+c$ has a formal power series solution in powers of $1/w$. This series is called the Bottcher coordinate. The $a_k$ are the coefficients of this power series. It converges in a neighborhood of $\infty$ and is a semi-conjugation from $x^2+c$ to the function $x^2$. $\endgroup$ – Joe Silverman Sep 16 '18 at 12:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.