:List of fractals by Hausdorff dimension

{{Short description|none}}

{{Use dmy dates|date=October 2019}}

According to Benoit Mandelbrot, "A fractal is by definition a set for which the Hausdorff-Besicovitch dimension strictly exceeds the topological dimension."{{harvnb|Mandelbrot|1982|p=15}}

Presented here is a list of fractals, ordered by increasing Hausdorff dimension, to illustrate what it means for a fractal to have a low or a high dimension.

Deterministic fractals

class="wikitable sortable"
Hausdorff dimension
(exact value) || Hausdorff dimension
(approx.) || Name || Illustration || width="40%" | Remarks
Calculatedalign="right" | 0.538Feigenbaum attractoralign="center" |150pxThe Feigenbaum attractor (see between arrows) is the set of points generated by successive iterations of the logistic map for the critical parameter value \lambda_\infty = 3.570, where the period doubling is infinite. This dimension is the same for any differentiable and unimodal function.{{cite journal |first=Erik |last=Aurell |title=On the metric properties of the Feigenbaum attractor |journal=Journal of Statistical Physics |volume=47 |issue=3–4 |pages=439–458 |date=May 1987 |doi=10.1007/BF01007519 |bibcode=1987JSP....47..439A |s2cid=122213380 }}
\log_3 2align="right" | 0.6309Cantor setalign="center" |200pxBuilt by removing the central third at each iteration. Nowhere dense and not a countable set.
\frac{-\log 2}{\log\left(\displaystyle\frac{1-\gamma}{2}\right)}align="right" | 0<D<11D generalized symmetric Cantor setalign="center" |200pxBuilt by removing the central interval of length \gamma\,l_{n-1} from each remaining interval of length l_{n-1} = (1-\gamma)^{n-1}/2^{n-1} at the nth iteration. \gamma=1/3 produces the usual middle-third Cantor set. Varying \gamma between 0 and 1 yields any fractal dimension 0.{{cite journal |first1=A. Yu |last1=Cherny |first2=E.M. |last2=Anitas |first3=A.I. |last3=Kuklin |first4=M. |last4=Balasoiu |first5=V.A. |last5=Osipov |title=The scattering from generalized Cantor fractals |journal=J. Appl. Crystallogr. |volume=43 |issue= 4|pages=790–7 |year=2010 |doi=10.1107/S0021889810014184 |arxiv=0911.2497 |s2cid=94779870 }}
\log_2 \varphi = \log_2(1+\sqrt{5}) - 1align="right" | 0.6942(1/4, 1/2) asymmetric Cantor setalign="center" |200pxBuilt by removing the second quarter at each iteration.{{Cite journal|author=Tsang, K. Y. |title=Dimensionality of Strange Attractors Determined Analytically |journal=Phys. Rev. Lett. |volume=57|issue=12|pages=1390–1393 |year=1986|pmid=10033437 |doi=10.1103/PhysRevLett.57.1390|bibcode=1986PhRvL..57.1390T}}

\varphi = (1+\sqrt{5})/2 (golden ratio).

\log_{10} 5 = 1 - \log_{10} 2align="right" | 0.69897Real numbers whose base 10 digits are evenalign="center" |200pxSimilar to the Cantor set.{{Cite book

| last = Falconer | first = Kenneth | author-link=Kenneth Falconer (mathematician)

| title = Fractal Geometry: Mathematical Foundations and Applications

| publisher = John Wiley & Sons, Ltd.

| year = 1990–2003

| isbn = 978-0-470-84862-3

| no-pp = true

| page = xxv}}

\log(1+\sqrt{2})align="right" | 0.88137Spectrum of Fibonacci Hamiltonianalign="center" |The study of the spectrum of the Fibonacci Hamiltonian proves upper and lower bounds for its fractal dimension in the large coupling regime. These bounds show that the spectrum converges to an explicit constant.{{cite journal |last1=Damanik |first1=D. |last2=Embree |first2=M. |last3=Gorodetski |first3=A. |first4=S. |last4=Tcheremchantse |title=The Fractal Dimension of the Spectrum of the Fibonacci Hamiltonian |journal=Commun. Math. Phys. |volume=280 |issue=2 |pages=499–516 |year=2008 |doi=10.1007/s00220-008-0451-3 |arxiv=0705.0338|bibcode=2008CMaPh.280..499D |s2cid=12245755 }}{{page needed|date=October 2018}}
1align="right" | 1Smith–Volterra–Cantor setalign="center" |200pxBuilt by removing the central interval of length 2^{-2n} from each remaining interval at the nth iteration. Nowhere dense but has a Lebesgue measure of 1/2.
2 + \log_2 \frac{1}{2} = 1align="right" | 1Takagi or Blancmange curvealign="center" |150pxDefined on the unit interval by f(x) = \sum\nolimits_{n=0}^\infty 2^{-n}s(2^{n}x), where s(x) is the triangle wave function. Not a fractal under Mandelbrot's definition, because its topological dimension is also 1.{{Cite book| last = Vaz| first = Cristina | title = Noções Elementares Sobre Dimensão | isbn = 9788565054867 | year = 2019}} Special case of the Takahi-Landsberg curve: f(x) = \sum\nolimits_{n=0}^\infty w^n s(2^n x) with w = 1/2. The Hausdorff dimension equals 2 + \log_2 w for w in \left[1/2,1\right]. (Hunt cited by Mandelbrot{{Cite book| last = Mandelbrot| first = Benoit | title = Gaussian self-affinity and Fractals | isbn = 978-0-387-98993-8 | year = 2002 | publisher = Springer }}).
Calculatedalign="right" | 1.0812Julia set z2 + 1/4align="center" |100pxJulia set of f(z) = z2 + 1/4.
Solution s of 2|\alpha|^{3s} + |\alpha|^{4s} = 1align="right" | 1.0933Boundary of the Rauzy fractalalign="center" |150pxFractal representation introduced by G.Rauzy of the dynamics associated to the Tribonacci morphism: 1\mapsto12, 2\mapsto13 and 3\mapsto1.Messaoudi, Ali. [http://matwbn.icm.edu.pl/ksiazki/aa/aa95/aa9531.pdf Frontième de numération complexe]", matwbn.icm.edu.pl. {{in lang|fr}} Accessed: 27 October 2018.{{page needed|date=October 2018}}{{Citation | last1=Lothaire | first1=M. | author-link=M. Lothaire | title=Applied combinatorics on words | url=https://archive.org/details/appliedcombinato0000loth/page/525 | publisher=Cambridge University Press | series=Encyclopedia of Mathematics and its Applications | isbn=978-0-521-84802-2 | mr=2165687 | zbl=1133.68067 | year=2005 | volume=105 | page=[https://archive.org/details/appliedcombinato0000loth/page/525 525] }} \alpha is one of the conjugated roots of z^3-z^2-z-1=0.
2\log_7 3align="right" | 1.12915contour of the Gosper islandalign="center" |100pxTerm used by Mandelbrot (1977).{{MathWorld |id=GosperIsland |title=Gosper Island |access-date=27 October 2018}} The Gosper island is the limit of the Gosper curve.
Measured (box counting)align="right" | 1.2Dendrite Julia setalign="center" |150pxJulia set of f(z) = z2 + i.
\frac{3\log \varphi}{\log\left(\displaystyle\frac{3+\sqrt{13}}{2}\right)}align="right" | 1.2083Fibonacci word fractal 60°align="center" | 200pxBuilt from the Fibonacci word. See also the standard Fibonacci word fractal.

\varphi = (1+\sqrt{5})/2 (golden ratio).

|\begin{align} & 2\log_2\left(\displaystyle\frac{\sqrt[3]{27-3\sqrt{78}}+\sqrt[3]{27+3\sqrt{78}}}{3}\right),\\ &\text{or root of }2^x-1=2^{(2-x)/2}\end{align}align="right" | 1.2108Boundary of the tame twindragonalign="center" |150pxOne of the six 2-rep-tiles in the plane (can be tiled by two copies of itself, of equal size).Ngai, Sirvent, Veerman, and Wang (October 2000). "[https://doi.org/10.1023%2FA%3A1005206301454 On 2-Reptiles in the Plane 1999]", Geometriae Dedicata, Volume 82. Accessed: 29 October 2018.Duda, Jarek (March 2011). "[http://demonstrations.wolfram.com/TheBoundaryOfPeriodicIteratedFunctionSystems/ The Boundary of Periodic Iterated Function Systems]", Wolfram.com.
align="right" | 1.26Hénon mapalign="center" |100pxThe canonical Hénon map (with parameters a = 1.4 and b = 0.3) has Hausdorff dimension 1.261 ± 0.003. Different parameters yield different dimension values.
\log_3 4align="right" | 1.2619Triflakealign="center" | 150pxThree anti-snowflakes arranged in a way that a koch-snowflake forms in between the anti-snowflakes.
\log_3 4align="right" | 1.2619Koch curvealign="center" | 200px3 Koch curves form the Koch snowflake or the anti-snowflake.
\log_3 4align="right" | 1.2619boundary of Terdragon curvealign="center" |150pxL-system: same as dragon curve with angle = 30°. The Fudgeflake is based on 3 initial segments placed in a triangle.
\log_3 4align="right" | 1.26192D Cantor dustalign="center" |100pxCantor set in 2 dimensions.
\log_3 4align="right" | 1.26192D L-system branchalign="center" |200pxL-Systems branching pattern having 4 new pieces scaled by 1/3. Generating the pattern using statistical instead of exact self-similarity yields the same fractal dimension.
Calculatedalign="right" | 1.2683Julia set z2 − 1align="center" |200pxJulia set of f(z) = z2 − 1.
align="right" | 1.3057Apollonian gasketalign="center" |120pxStarting with 3 tangent circles, repeatedly packing new circles into the complementary interstices. Also the limit set generated by reflections in 4 mutually tangent circles. See McMullen, Curtis T. (3 October 1997). "[http://abel.math.harvard.edu/~ctm/papers/home/text/papers/dimIII/dimIII.pdf Hausdorff dimension and conformal dynamics III: Computation of dimension]", Abel.Math.Harvard.edu. Accessed: 27 October 2018.
align="right" | 1.3285 circles inversion fractalalign="center" |100pxThe limit set generated by iterated inversions with respect to 5 mutually tangent circles (in red). Also an Apollonian packing. See Chang, Angel and Zhang, Tianrong. {{Cite web |url=http://classes.yale.edu/fractals/CircInvFrac/CircDim/CircDim2.html |title=On the Fractal Structure of the Boundary of Dragon Curve |access-date=9 February 2019 |archive-url=https://web.archive.org/web/20110614063904/http://classes.yale.edu/Fractals/CircInvFrac/CircDim/CircDim2.html |archive-date=14 June 2011 |url-status=bot: unknown }} [https://stanford.edu/~angelx/pubs/dragonbound.pdf pdf]
\log_5 9align="right" | 1.36521Mandelbrot, B. B. (1983). The Fractal Geometry of Nature, p.48. New York: W. H. Freeman. {{ISBN|9780716711865}}. Cited in: {{MathWorld |id=MinkowskiSausage |title=Minkowski Sausage |access-date=22 September 2019}}Quadratic von Koch island using the type 1 curve as generatoralign="center" |150pxAlso known as the Minkowski Sausage
Calculatedalign="right" | 1.3934Douady rabbitalign="center" |150pxJulia set of f(z) = -0.123 + 0.745i
\log_3 5align="right" | 1.4649Vicsek fractalalign="center" |100pxBuilt by exchanging iteratively each square by a cross of 5 squares.
\log_3 5align="right" | 1.4649Quadratic von Koch curve (type 1)align="center" |150pxOne can recognize the pattern of the Vicsek fractal (above).
\log_{\sqrt{5}} \frac{10}{3}align="right" | 1.4961Quadric crossalign="center" |{{anchor|cross}}150pxFile:Q Cross Fractal Generator.jpgBuilt by replacing each end segment with a cross segment scaled by a factor of 51/2, consisting of 3 1/3 new segments, as illustrated in the inset.

Images generated with Fractal Generator for ImageJ.

2 - \log_2 \sqrt{2} = \frac{3}{2}align="right" | 1.5000a Weierstrass function: f(x) = \sum_{k=1}^\infty \frac{\sin(2^k x)}{\sqrt{2}^{\,k}}align="center" |150pxThe Hausdorff dimension of the graph of the Weierstrass function f:[0,1] \to \mathbb{R} defined by f(x) = \sum\nolimits_{k=1}^\infty a^k\sin(b^k x) with 1/b < a < 1 and b>1 is 2 + \log_b a.{{Cite journal|last=Shen|first=Weixiao|date=2018|title=Hausdorff dimension of the graphs of the classical Weierstrass functions|journal=Mathematische Zeitschrift|language=en|volume=289|issue=1–2|pages=223–266|doi=10.1007/s00209-017-1949-1|issn=0025-5874|arxiv=1505.03986|s2cid=118844077}}N. Zhang. The Hausdorff dimension of the graphs of fractal functions. (In Chinese). Master Thesis. Zhejiang University, 2018.
\log_4 8 = \frac{3}{2}align="right" | 1.5000Quadratic von Koch curve (type 2)align="center" |150pxAlso called "Minkowski sausage".
\log_2\left(\frac{1+\sqrt[3]{73-6\sqrt{87}}+\sqrt[3]{73+6\sqrt{87}}}{3}\right)align="right" | 1.5236Boundary of the Dragon curvealign="center" | 150pxcf. Chang & Zhang.[http://poignance.coiraweb.com/math/Fractals/Dragon/Bound.html Fractal dimension of the boundary of the dragon fractal]
\log_2\left(\frac{1+\sqrt[3]{73-6\sqrt{87}}+\sqrt[3]{73+6\sqrt{87}}}{3}\right)align="right" | 1.5236Boundary of the twindragon curvealign="center" |150pxCan be built with two dragon curves. One of the six 2-rep-tiles in the plane (can be tiled by two copies of itself, of equal size).
\log_2 3align="right" | 1.58503-branches treealign="center" | 110px 110pxEach branch carries 3 branches (here 90° and 60°). The fractal dimension of the entire tree is the fractal dimension of the terminal branches. NB: the 2-branches tree has a fractal dimension of only 1.
\log_2 3align="right" | 1.5850Sierpinski trianglealign="center" | 100pxAlso the limiting shape of Pascal's triangle modulo 2.
\log_2 3align="right" | 1.5850Sierpiński arrowhead curvealign="center" | 100pxSame limit as the triangle (above) but built with a one-dimensional curve.
\log_2 3align="right" | 1.5850Boundary of the T-square fractalalign="center" | 200pxThe dimension of the fractal itself (not the boundary) is \log_2 4 = 2
\log_{\sqrt[\varphi]{\varphi}}(\varphi) = \varphialign="right" | 1.61803a golden dragonalign="center" | 150pxBuilt from two similarities of ratios r and r^2, with r = 1 / \varphi^{1/\varphi}. Its dimension equals \varphi because ({r^2})^\varphi + r^\varphi = 1.

\varphi = (1+\sqrt{5})/2 (golden ratio).

1 + \log_3 2align="right" | 1.6309Pascal triangle modulo 3align="center" | 160pxFor a triangle modulo k, if k is prime, the fractal dimension is 1 + \log_k\left(\frac{k+1}{2}\right) (cf. Stephen Wolfram{{Cite web |url=http://www.stephenwolfram.com/publications/articles/ca/84-geometry/1/text.html |title=Fractal dimension of the Pascal triangle modulo k |access-date=2 October 2006 |archive-date=15 October 2012 |archive-url=https://web.archive.org/web/20121015154035/http://www.stephenwolfram.com/publications/articles/ca/84-geometry/1/text.html |url-status=dead }}).
1 + \log_3 2align="right" | 1.6309Sierpinski Hexagonalign="center" | 150pxBuilt in the manner of the Sierpinski carpet, on an hexagonal grid, with 6 similitudes of ratio 1/3. The Koch snowflake is present at all scales.
\frac{3\log \varphi}{\log(1+\sqrt{2})}align="right" | 1.6379Fibonacci word fractalalign="center" | 150pxFractal based on the Fibonacci word (or Rabbit sequence) Sloane A005614. Illustration : Fractal curve after 23 steps (F23 = 28657 segments).[http://hal.archives-ouvertes.fr/hal-00367972/en/ The Fibonacci word fractal]

\varphi = (1+\sqrt{5})/2 (golden ratio).

Solution of (1/3)^s + (1/2)^s + (2/3)^s = 1align="right" | 1.6402Attractor of IFS with 3 similarities of ratios 1/3, 1/2 and 2/3align="center" | 200pxGeneralization : Providing the open set condition holds, the attractor of an iterated function system consisting of n similarities of ratios c_n, has Hausdorff dimension s, solution of the equation coinciding with the iteration function of the Euclidean contraction factor: \sum\nolimits_{k=1}^n c_k^s = 1.
\log_8 32 = \frac{5}{3}align="right" | 1.666732-segment quadric fractal (1/8 scaling rule)frameless see also: :File:32 Segment One Eighth Scale Quadric Fractal.jpgFile:32SegmentSmall.jpgBuilt by scaling the 32 segment generator (see inset) by 1/8 for each iteration, and replacing each segment of the previous structure with a scaled copy of the entire generator. The structure shown is made of 4 generator units and is iterated 3 times. The fractal dimension for the theoretical structure is log 32/log 8 = 1.6667. Images generated with Fractal Generator for ImageJ.
1 + \log_5 3align="right" | 1.6826Pascal triangle modulo 5align="center" | 160pxFor a triangle modulo k, if k is prime, the fractal dimension is 1 + \log_k\left(\frac{k+1}{2}\right) (cf. Stephen Wolfram).
Measured (box-counting)align="right" | 1.7Ikeda map attractoralign="center" | 100pxFor parameters a=1, b=0.9, k=0.4 and p=6 in the Ikeda map z_{n+1} = a + bz_n \exp\left[i\left[k - p/\left(1 + \lfloor z_n \rfloor^2\right)\right]\right]. It derives from a model of the plane-wave interactivity field in an optical ring laser. Different parameters yield different values.{{cite journal |first=James |last=Theiler |title=Estimating fractal dimension |journal=J. Opt. Soc. Am. A |volume=7 |issue=6 |pages=1055–73 |year=1990 |doi=10.1364/JOSAA.7.001055 |bibcode=1990JOSAA...7.1055T |url=http://public.lanl.gov/jt/Papers/est-fractal-dim.pdf }}
1 + \log_{10} 5align="right" | 1.699050 segment quadric fractal (1/10 scaling rule)align="center" | 150pxBuilt by scaling the 50 segment generator (see inset) by 1/10 for each iteration, and replacing each segment of the previous structure with a scaled copy of the entire generator. The structure shown is made of 4 generator units and is iterated 3 times. The fractal dimension for the theoretical structure is log 50/log 10 = 1.6990. Images generated with Fractal Generator for ImageJ[http://rsb.info.nih.gov/ij/plugins/fractal-generator.html Fractal Generator for ImageJ] {{webarchive|url=https://web.archive.org/web/20120320124725/http://rsb.info.nih.gov/ij/plugins/fractal-generator.html |date=20 March 2012 }}..File:50SegmentSmall.jpg
4\log_5 2align="right" | 1.7227Pinwheel fractalalign="center" | 150pxBuilt with Conway's Pinwheel tile.
\log_3 7align="right" | 1.7712Sphinx fractalalign="center" | 150pxBuilt with the Sphinx hexiamond tiling, removing two of the nine sub-sphinxes.W. Trump, G. Huber, C. Knecht, R. Ziff, to be published
\log_3 7align="right" | 1.7712Hexaflakealign="center" | 100pxBuilt by exchanging iteratively each hexagon by a flake of 7 hexagons. Its boundary is the von Koch flake and contains an infinity of Koch snowflakes (black or white).
\log_3 7align="right" | 1.7712Fractal H-I de Riveraalign="center" | 100pxStarting from a unit square dividing its dimensions into three equal parts to form nine self-similar squares with the first square, two middle squares (the one that is above and the one below the central square) are removed in each of the seven squares not eliminated the process is repeated, so it continues indefinitely.
\frac{\log 4}{\log(2 + 2\cos(85^\circ))}align="right" | 1.7848Von Koch curve 85°align="center" | 150pxGeneralizing the von Koch curve with an angle a chosen between 0 and 90°. The fractal dimension is then \frac{\log 4}{\log(2 + 2\cos a)} \in [1,2].
\log_2\left(3^{0.63} + 2^{0.63}\right)align="right" | 1.8272A self-affine fractal setalign="center" | 200pxBuild iteratively from a p-by-q array on a square, with p \le q. Its Hausdorff dimension equals \log_p\left(\sum\nolimits_{k=1}^p n_k^a\right) with a = \log_q p and n_k is the number of elements in the kth column. The box-counting dimension yields a different formula, therefore, a different value. Unlike self-similar sets, the Hausdorff dimension of self-affine sets depends on the position of the iterated elements and there is no formula, so far, for the general case.
\frac{\log 6}{\log(1+\varphi)}align="right" | 1.8617Pentaflakealign="center" | 100pxBuilt by exchanging iteratively each pentagon by a flake of 6 pentagons.

\varphi = (1+\sqrt{5})/2 (golden ratio).

solution of 6(1/3)^s + 5{(1/3\sqrt{3})}^s = 1align="right" | 1.8687Monkeys treealign="center" | 100pxThis curve appeared in Benoit Mandelbrot's "Fractal geometry of Nature" (1983). It is based on 6 similarities of ratio 1/3 and 5 similarities of ratio 1/3\sqrt{3}.[http://www.coaauw.org/boulder-eyh/eyh_fractal.html Monkeys tree fractal curve] {{webarchive|url=https://archive.today/20020921135308/http://www.coaauw.org/boulder-eyh/eyh_fractal.html |date=21 September 2002 }}
\log_3 8align="right" | 1.8928Sierpinski carpetalign="center" | 100pxEach face of the Menger sponge is a Sierpinski carpet, as is the bottom surface of the 3D quadratic Koch surface (type 1).
\log_3 8align="right" | 1.89283D Cantor dustalign="center" | 200pxCantor set in 3 dimensions.
\log_3 4 + \log_3 2 = \frac{\log 4}{\log 3}+\frac{\log 2}{\log 3} = \frac{\log 8}{\log 3}align="right" | {{formatnum:1.8928}}Cartesian product of the von Koch curve and the Cantor setalign="center" | 150pxGeneralization : Let F×G be the cartesian product of two fractals sets F and G. Then \dim_H(F \times G) = \dim_H F + \dim_H G. See also the 2D Cantor dust and the Cantor cube.
2\log_2 x where x^9-3x^8+3x^7-3x^6+2x^5+4x^4-8x^3+8x^2-16x+8 = 0align="right" | 1.9340Boundary of the Lévy C curvealign="center" | 100pxEstimated by Duvall and Keesling (1999). The curve itself has a fractal dimension of 2.
align="right" | 2Penrose tilingalign="center" |100pxSee Ramachandrarao, Sinha & Sanyal.[http://www.iisc.ernet.in/currsci/aug102000/rc80.pdf Fractal dimension of a Penrose tiling]
2align="right" | 2Boundary of the Mandelbrot setalign="center" | 100pxThe boundary and the set itself have the same Hausdorff dimension.{{cite arXiv |first=Mitsuhiro |last=Shishikura |title=The Hausdorff dimension of the boundary of the Mandelbrot set and Julia sets |date=1991 |eprint=math/9201282}}
2align="right" | 2Julia setalign="center" | 150pxFor determined values of c (including c belonging to the boundary of the Mandelbrot set), the Julia set has a dimension of 2.
2align="right" | 2Sierpiński curvealign="center" | 100pxEvery space-filling curve filling the plane has a Hausdorff dimension of 2.
2align="right" | 2Hilbert curvealign="center" | 100px
2align="right" | 2Peano curvealign="center" | 100pxAnd a family of curves built in a similar way, such as the Wunderlich curves.
2align="right" | 2Moore curvealign="center" | 100pxCan be extended in 3 dimensions.
align="right" | 2Lebesgue curve or z-order curvealign="center" | 100pxUnlike the previous ones this space-filling curve is differentiable almost everywhere. Another type can be defined in 2D. Like the Hilbert Curve it can be extended in 3D.[http://www.mathcurve.com/fractals/lebesgue/lebesgue.shtml Lebesgue curve variants]
\log_{\sqrt{2}} 2 = 2align="right" | 2Dragon curvealign="center" | 150pxAnd its boundary has a fractal dimension of 1.5236270862.{{cite arXiv |first=Jarek |last=Duda |title=Complex base numeral systems |date=2008 |class=math.DS |eprint=0712.1309v3}}
align="right" | 2Terdragon curvealign="center" | 150pxL-system: F → F + F – F, angle = 120°.
\log_2 4 = 2align="right" | 2Gosper curvealign="center" | 100pxIts boundary is the Gosper island.
Solution of 7({1/3})^s+6({1/3\sqrt{3}})^s=1align="right" | 2Curve filling the Koch snowflakealign="center" | 100pxProposed by Mandelbrot in 1982,{{cite book |title=Penser les mathématiques |last=Seuil |isbn=2-02-006061-2 |year=1982|publisher=Seuil }} it fills the Koch snowflake. It is based on 7 similarities of ratio 1/3 and 6 similarities of ratio 1/3\sqrt{3}.
\log_2 4 = 2align="right" | 2Sierpiński tetrahedronalign="center" | 80pxEach tetrahedron is replaced by 4 tetrahedra.
\log_2 4 = 2align="right" | 2H-fractalalign="center" |150pxAlso the Mandelbrot tree which has a similar pattern.
\frac{\log 2}{\log(2/\sqrt{2})} = 2align="right" | {{formatnum:2}}Pythagoras tree (fractal)align="center" |150pxEvery square generates two squares with a reduction ratio of 1/\sqrt{2}.
\log_2 4 = 2align="right" | 22D Greek cross fractalalign="center" |100pxEach segment is replaced by a cross formed by 4 segments.
Measuredalign="right" | 2.01 ± 0.01Rössler attractoralign="center" | 100pxThe fractal dimension of the Rössler attractor is slightly above 2. For a=0.1, b=0.1 and c=14 it has been estimated between 2.01 and 2.02.[http://www.ocf.berkeley.edu/~trose/rossler.html Fractals and the Rössler attractor]
Measuredalign="right" | 2.06 ± 0.01Lorenz attractoralign="center" |100pxFor parameters \rho=40, \sigma=16 and \beta=4 . See McGuinness (1983){{cite journal |first=M.J. |last=McGuinness |title=The fractal dimension of the Lorenz attractor |journal=Physics Letters |volume=99A |pages=5–9 |year=1983 |issue=1 |doi=10.1016/0375-9601(83)90052-X |bibcode=1983PhLA...99....5M }}
4 + c^D + d^D = (c+d)^Dalign="right" |2<D<2.3Pyramid surfacealign="center" |200pxEach triangle is replaced by 6 triangles, of which 4 identical triangles form a diamond based pyramid and the remaining two remain flat with lengths c and d relative to the pyramid triangles. The dimension is a parameter, self-intersection occurs for values greater than 2.3.{{Cite journal|last=Lowe|first=Thomas|date=24 October 2016|title=Three Variable Dimension Surfaces|url=https://www.researchgate.net/publication/309391846|journal=ResearchGate}}
\log_2 5align="right" | 2.3219Fractal pyramidalign="center" |100pxEach square pyramid is replaced by 5 half-size square pyramids. (Different from the Sierpinski tetrahedron, which replaces each triangular pyramid with 4 half-size triangular pyramids).
\frac{\log 20}{\log(2+\varphi)}align="right" | 2.3296Dodecahedron fractalalign="center" |100pxEach dodecahedron is replaced by 20 dodecahedra.

\varphi = (1+\sqrt{5})/2 (golden ratio).

\log_3 13align="right" | 2.33473D quadratic Koch surface (type 1)align="center" |150pxExtension in 3D of the quadratic Koch curve (type 1). The illustration shows the first (blue block), second (plus green blocks), third (plus yellow blocks) and fourth (plus clear blocks) iterations.
align="right" | 2.4739Apollonian sphere packingalign="center" |100pxThe interstice left by the Apollonian spheres. Apollonian gasket in 3D. Dimension calculated by M. Borkovec, W. De Paris, and R. Peikert.[http://www.scivis.ethz.ch/publications/pdf/1994/borkovec1994fractal.pdf The Fractal dimension of the apollonian sphere packing] {{webarchive|url=https://web.archive.org/web/20160506190118/http://www.scivis.ethz.ch/publications/pdf/1994/borkovec1994fractal.pdf |date=6 May 2016 }}
\log_4 32 = \frac{5}{2}align="right" | 2.503D quadratic Koch surface (type 2)align="center" |150pxExtension in 3D of the quadratic Koch curve (type 2). The illustration shows the second iteration.
\frac{\log\left(\frac{\sqrt{7}}{6}-\frac{1}{3}\right)}{\log(\sqrt2-1)}align="right" | 2.529Jerusalem cubealign="center" | 150pxThe iteration n is built with 8 cubes of iteration n−1 (at the corners) and 12 cubes of iteration n-2 (linking the corners). The contraction ratio is \sqrt{2}-1.
\frac{\log 12}{\log(1+\varphi)}align="right" | 2.5819Icosahedron fractalalign="center" |100pxEach icosahedron is replaced by 12 icosahedra.

\varphi = (1+\sqrt{5})/2 (golden ratio).

1 + \log_2 3align="right" | 2.58493D Greek cross fractalalign="center" |200pxEach segment is replaced by a cross formed by 6 segments.
1 + \log_2 3align="right" | 2.5849Octahedron fractalalign="center" |100pxEach octahedron is replaced by 6 octahedra.
1 + \log_2 3align="right" | 2.5849von Koch surfacealign="center" |150pxEach equilateral triangular face is cut into 4 equal triangles.

Using the central triangle as the base, form a tetrahedron. Replace the triangular base with the tetrahedral "tent".

\frac{\log 3}{\log(3/2)}align="right" | 2.7095Von Koch in 3Dalign="center" | 100pxStart with a 6-sided polyhedron whose faces are isosceles triangles with sides of ratio 2:2:3 . Replace each polyhedron with 3 copies of itself, 2/3 smaller.{{cite web |url=https://www.researchgate.net/publication/262600735 |last=Baird |first=Eric |date=2014 |title=The Koch curve in three dimensions |via=ResearchGate}}
\log_3 20align="right" | 2.7268Menger spongealign="center" | 100pxAnd its surface has a fractal dimension of \log_3 20, which is the same as that by volume.
\log_2 8 = 3align="right" | 33D Hilbert curvealign="center" | 100pxA Hilbert curve extended to 3 dimensions.
\log_2 8 = 3align="right" | 33D Lebesgue curvealign="center" | 100pxA Lebesgue curve extended to 3 dimensions.
\log_2 8 = 3align="right" | 33D Moore curvealign="center" | 100pxA Moore curve extended to 3 dimensions.
\log_2 8 = 3align="right" | 33D H-fractalalign="center" | 120pxA H-fractal extended to 3 dimensions.{{cite journal |first1=B. |last1=Hou |first2=H. |last2=Xie |first3=W. |last3=Wen |first4=P. |last4=Sheng | year = 2008

| title = Three-dimensional metallic fractals and their photonic crystal characteristics |journal= Phys. Rev. B |volume=77 |issue=12 |page=125113

|bibcode=2008PhRvB..77l5113H |doi=10.1103/PhysRevB.77.125113 |url=http://repository.ust.hk/ir/bitstream/1783.1-25969/1/PhysRevB.77.125113.pdf }}

3 (conjectured)align="right" | {{formatnum:3}} (to be confirmed)Mandelbulbalign="center" |100pxExtension of the Mandelbrot set (power 9) in 3 dimensions[http://www.fractalforums.com/theory/hausdorff-dimension-of-the-mandelbulb/15/ Hausdorff dimension of the Mandelbulb]{{Unreliable source?|date=September 2011}}

Random and natural fractals

class="wikitable sortable"
Hausdorff dimension
(exact value) || Hausdorff dimension
(approx.) || Name || Illustration || width="40%" | Remarks
\frac{1}{2}align="right" | 0.5Zeros of a Wiener processalign="center" |150pxThe zeros of a Wiener process (Brownian motion) are a nowhere dense set of Lebesgue measure 0 with a fractal structure.Peter Mörters, Yuval Peres, "Brownian Motion", Cambridge University Press, 2010
Solution of E(C_1^s + C_2^s)=1 where E(C_1)=0.5 and E(C_2)=0.3align="right" | 0.7499a random Cantor set with 50% - 30%align="center" |150pxGeneralization: at each iteration, the length of the left interval is defined with a random variable C_1, a variable percentage of the length of the original interval. Same for the right interval, with a random variable C_2. Its Hausdorff Dimension s satisfies: E(C_1^s + C_2^s)=1 (where E(X) is the expected value of X).
Solution of s+1 = 12\cdot2^{-(s+1)}-6\cdot3^{-(s+1)}align="right"|1.144...von Koch curve with random intervalalign="center"| 200pxThe length of the middle interval is a random variable with uniform distribution on the interval (0,1/3).
Measuredalign="right"|1.22 ± 0.02Coastline of Irelandalign="center"| 150pxValues for the fractal dimension of the entire coast of Ireland were determined by McCartney, Abernethy and Gault{{cite journal|last1=McCartney|first1=Mark|first2=Gavin |last2=Abernethya |first3=Lisa |last3=Gaulta|title=The Divider Dimension of the Irish Coast|journal=Irish Geography|date=24 June 2010|volume=43|issue=3|pages=277–284|doi=10.1080/00750778.2011.582632}} at the University of Ulster and Theoretical Physics students at Trinity College, Dublin, under the supervision of S. Hutzler.{{cite journal|last1=Hutzler |first1=S. |title=Fractal Ireland |journal=Science Spin |date=2013 |volume=58 |pages=19–20 |url=https://issuu.com/spin35/docs/spin_58_all |access-date=15 November 2016 }}

(See [https://web.archive.org/web/20130726164417/http://www.sciencespin.com/magazine/archive/2013/05/ contents page], archived 26 July 2013)

Note that there are marked differences between Ireland's ragged west coast (fractal dimension of about 1.26) and the much smoother east coast (fractal dimension 1.10)

Measuredalign="right"|1.25Coastline of Great Britainalign="center"| 200pxFractal dimension of the west coast of Great Britain, as measured by Lewis Fry Richardson and cited by Benoît Mandelbrot.[http://users.math.yale.edu/~bbm3/web_pdfs/howLongIsTheCoastOfBritain.pdf How long is the coast of Britain? Statistical self-similarity and fractional dimension], B. Mandelbrot
\frac{\log 4}{\log 3}align="right" | 1.2619von Koch curve with random orientationalign="center" | 200pxOne introduces here an element of randomness which does not affect the dimension, by choosing, at each iteration, to place the equilateral triangle above or below the curve.
\frac{4}{3}align="right" | 1.333Boundary of Brownian motionalign="center" |150px(cf. Mandelbrot, Lawler, Schramm, Werner).{{cite journal |first1=Gregory F. |last1=Lawler |first2=Oded |last2=Schramm |first3=Wendelin |last3=Werner |title=The Dimension of the Planar Brownian Frontier is 4/3 |journal=Math. Res. Lett. |volume=8 |issue=4 |pages=401–411 |year=2001 |arxiv=math/0010165|bibcode=2000math.....10165L |doi=10.4310/MRL.2001.v8.n4.a1 |s2cid=5877745 }}
\frac{4}{3}align="right" | 1.333Polymer in 2Dalign="center" |Similar to the Brownian motion in 2D with non-self-intersection.{{cite book |first=Bernard |last=Sapoval |title=Universalités et fractales |publisher=Flammarion-Champs |year=2001 |isbn=2-08-081466-4 }}
\frac{4}{3}align="right" | 1.333Percolation front in 2D, Corrosion front in 2Dalign="center" | 150pxFractal dimension of the percolation-by-invasion front (accessible perimeter), at the percolation threshold (59.3%). It's also the fractal dimension of a stopped corrosion front.
align="right" | 1.40Clusters of clusters 2Dalign="center" |When limited by diffusion, clusters combine progressively to a unique cluster of dimension 1.4.
2-\frac{1}{2}align="right" | 1.5Graph of a regular Brownian function (Wiener process)align="center" | 150pxGraph of a function f such that, for any two positive reals x and x+h, the difference of their images f(x+h)-f(x) has the centered gaussian distribution with variance h. Generalization: the fractional Brownian motion of index \alpha follows the same definition but with a variance h^{2\alpha}, in that case its Hausdorff dimension equals 2-\alpha.
Measuredalign="right" | 1.52Coastline of Norwayalign="center" |100pxSee J. Feder.Feder, J., "Fractals", Plenum Press, New York, (1988).
Measuredalign="right" | 1.55Self-avoiding walkalign="center" | 150pxRandom walk in a square lattice that avoids visiting the same place twice, with a "go-back" routine for avoiding dead ends.
\frac{5}{3}align="right" | 1.66Polymer in 3Dalign="center" |Similar to the Brownian motion in a cubic lattice, but without self-intersection.
align="right" | 1.702D DLA Clusteralign="center" | 150pxIn 2 dimensions, clusters formed by diffusion-limited aggregation, have a fractal dimension of around 1.70.
\frac{\log(9\cdot0.75)}{\log 3}align="right" | 1.7381Fractal percolation with 75% probabilityalign="center" |150pxThe fractal percolation model is constructed by the progressive replacement of each square by a 3-by-3 grid in which is placed a random collection of sub-squares, each sub-square being retained with probability p. The "almost sure" Hausdorff dimension equals \frac{\log(9p)}{\log 3}.
\frac{7}{4}align="right" | 1.752D percolation cluster hullalign="center" | 150pxThe hull or boundary of a percolation cluster. Can also be generated by a hull-generating walk,[http://deepblue.lib.umich.edu/handle/2027.42/27787 Hull-generating walks] or by Schramm-Loewner Evolution.
\frac{91}{48}align="right" | 1.89582D percolation clusteralign="center" | 150pxIn a square lattice, under the site percolation threshold (59.3%) the percolation-by-invasion cluster has a fractal dimension of 91/48.{{cite book|author1=M Sahini|author2=M Sahimi|title=Applications Of Percolation Theory|url=https://books.google.com/books?id=MJwqsbWBc-YC|year=2003|publisher=CRC Press|isbn=978-0-203-22153-2}} Beyond that threshold, the cluster is infinite and 91/48 becomes the fractal dimension of the "clearings".
\frac{\log 2}{\log \sqrt{2}} = 2align="right" | 2Brownian motionalign="center" | 150pxOr random walk. The Hausdorff dimensions equals 2 in 2D, in 3D and in all greater dimensions (K.Falconer "The geometry of fractal sets").
Measuredalign="right" | Around 2Distribution of galaxy clustersalign="center" | 100pxFrom the 2005 results of the Sloan Digital Sky Survey.[https://arxiv.org/abs/astro-ph/0501583v2 Basic properties of galaxy clustering in the light of recent results from the Sloan Digital Sky Survey]
align="right" | 2.5Balls of crumpled paperalign="center" | 100pxWhen crumpling sheets of different sizes but made of the same type of paper and with the same aspect ratio (for example, different sizes in the ISO 216 A series), then the diameter of the balls so obtained elevated to a non-integer exponent between 2 and 3 will be approximately proportional to the area of the sheets from which the balls have been made.{{Cite journal | publisher=Yale | url=http://classes.yale.edu/fractals/FracAndDim/BoxDim/PowerLaw/CrumpledPaper.html | title=Power Law Relations | access-date=29 July 2010 | url-status=dead | archive-url=https://web.archive.org/web/20100628020140/http://classes.yale.edu/fractals/FracAndDim/BoxDim/PowerLaw/CrumpledPaper.html | archive-date=28 June 2010 | df=dmy-all }} Creases will form at all size scales (see Universality (dynamical systems)).
align="right" | 2.503D DLA Clusteralign="center" | 150pxIn 3 dimensions, clusters formed by diffusion-limited aggregation, have a fractal dimension of around 2.50.
align="right" | 2.50Lichtenberg figurealign="center" | 100pxTheir appearance and growth appear to be related to the process of diffusion-limited aggregation or DLA.
3-\frac{1}{2}align="right" | 2.5regular Brownian surfacealign="center" | 150pxA function f:\mathbb{R}^2 \to \mathbb{R}, gives the height of a point (x,y) such that, for two given positive increments h and k, then f(x+h,y+k)-f(x,y) has a centered Gaussian distribution with variance \sqrt{h^2 + k^2}. Generalization: the fractional Brownian surface of index \alpha follows the same definition but with a variance (h^2+k^2)^\alpha, in that case its Hausdorff dimension equals 3-\alpha.
Measuredalign="right" | 2.523D percolation clusteralign="center" |225pxIn a cubic lattice, at the site percolation threshold (31.1%), the 3D percolation-by-invasion cluster has a fractal dimension of around 2.52. Beyond that threshold, the cluster is infinite.
Measured and calculatedalign="right" |~2.7The surface of Broccolialign="center" | 100pxSan-Hoon Kim used a direct scanning method and a cross section analysis of a broccoli to conclude that the fractal dimension of it is ~2.7.{{Cite arXiv |eprint=cond-mat/0411597|title=Fractal dimensions of a green broccoli and a white cauliflower|last=Kim|first=Sang-Hoon|date=2 February 2008}}
Measuredalign="right" | ~2.8Surface of human brainalign="center" | 100pxMeasured with segmented three-dimensional high-resolution magnetic resonance images{{cite journal |first1=Valerij G. |last1=Kiselev |first2=Klaus R. |last2=Hahn |first3=Dorothee P. |last3=Auer |title=Is the brain cortex a fractal? |journal=NeuroImage |volume=20 |issue=3 |pages=1765–1774 |year=2003 |doi=10.1016/S1053-8119(03)00380-X|pmid=14642486 |s2cid=14240006 }}
Measured and calculatedalign="right" |~2.8Caulifloweralign="center" | 100pxSan-Hoon Kim used a direct scanning method and a mathematical analysis of the cross section of a cauliflower to conclude that the fractal dimension of it is ~2.8.
align="right" | 2.97Lung surfacealign="center" |100pxThe alveoli of a lung form a fractal surface close to 3.
Calculatedalign="right" | \in (0,2)Multiplicative cascadealign="center" | 150pxThis is an example of a multifractal distribution. However, by choosing its parameters in a particular way we can force the distribution to become a monofractal.{{cite journal |first1=Paul |last1=Meakin |title=Diffusion-limited aggregation on multifractal lattices: A model for fluid-fluid displacement in porous media |journal=Physical Review A |volume=36 |issue=6 |pages=2833–2837 |year=1987 |doi=10.1103/PhysRevA.36.2833|pmid=9899187 |bibcode=1987PhRvA..36.2833M }}

See also

Notes and references

{{Reflist|30em}}

Further reading

  • {{cite book |first=Benoît |last=Mandelbrot |title=The Fractal Geometry of Nature |publisher=W.H. Freeman |year=1982 |isbn=0-7167-1186-9 |url-access=registration |url=https://archive.org/details/fractalgeometryo00beno }}
  • {{cite book |first=Heinz-Otto |last=Peitgen |editor-first=Dietmar |editor-last=Saupe |title=The Science of Fractal Images |publisher=Springer Verlag |year=1988 |isbn=0-387-96608-0 |url-access=registration |url=https://archive.org/details/scienceoffractal0000unse }}
  • {{cite book |first=Michael F. |last=Barnsley |title=Fractals Everywhere |publisher=Morgan Kaufmann |isbn=0-12-079061-0 |date=1 January 1993 }}
  • {{cite book |first1=Bernard |last1=Sapoval |first2=Benoît B. |last2=Mandelbrot |title=Universalités et fractales: jeux d'enfant ou délits d'initié? |publisher=Flammarion-Champs |year=2001 |isbn=2-08-081466-4 }}