On the Biplanarity of Blowups

David Eppstein

Graph Drawing 2023

 Ringel’s Earth–Moon problem (1959)

Countries build moon colonies,

with different borders than Earth

We must color the maps of the

Earth and Moon, so
▶ Each country gets a single
color for both maps
▶ Adjacent countries on either
map get different colors
▶ We use few colors

How many colors do we need?

 Earth–Moon maps as graph drawings
Two equivalent versions:
▶ Draw graph as two planar subgraphs, covering all edges, in two separate planes
(the dual graphs of the countries)
▶ Draw graph in one plane, with two colors of edges, so that no two edges of the
same color cross (may need curved edges)
a f a f a f

b b b

c e c e c e

i i i

d d d

h j h j h j

k k k

g l g l g l

# separate planes = # edge colors for no monochrome crossing = thickness

Graph is biplanar if its thickness is 2, or equivalently if it is dual to an Earth–Moon map

 Sulanke’s 9-chromatic Earth–Moon map (1974)
Connect all vertices of a 5-cycle to all vertices of a 6-clique
A F

B E
C D Q
P A T

P F
B C
C
Q S T D
R
D F E S A
E

R B R
Q S

P T

Requires 9 colors — 3 for cycle, 6 for clique

Best upper bound for arbitrary biplanar graphs is 12 colors

Kicked off a line of research: which simple combinations of smaller graphs are biplanar?
 Blowups

k-blowup: replace every vertex by a k-vertex independent set

Replace every edge by a complete bipartite subgraph connecting two independent sets

Conjecture [Gethner 2018]: 2-blowups of planar graphs are biplanar

 Weak evidence for the conjecture
Count the edges!

A planar graph with n vertices has ≤ 3n − 6 edges

Its blowup has 2n vertices and ≤ 12n − 24 edges

But a biplanar graph with 2n vertices can have up to 12n − 12 edges, twelve more edges
 When counting is strong enough
Hereditary family with ≤ 2n − 2 edges ⇒ cover edges by two trees [Nash-Williams 1964]
2-blowup of a tree is planar ⇒ 2-blowup of whole graph is biplanar [Albertson et al. 2010]
a b
a b
b b e e
a b

e g
c d e d g g d d f f d
e g

f c
c c a a
f g
f c

Works for triangle-free planar (≤ 2n − 4 edges) but not maximal planar (3n − 6 edges)
Proves 2-blowup thickness ≤ 3 for all planar graphs
Some maximal planar graphs with biplanar blowups

Partition dual graph into two induced paths ⇒ outerpaths in primal

a
a
l b c d e
d c
e k j i h f
b
l b g
h i
c a l k j
j

f
k
d e f g i
g
h
 Some maximal planar graphs with biplanar blowups
For each outerpath, draw all four copies of each diagonal edge and two of the four
copies of each boundary edge as a planar graph
a0 b0

l0 b0 c1 a1
k0 c0 d0 l0
j0 d0 e1 k1
i0 e0 f0 j0
h0 f0 g1 i1
g0 h0

g1 h1
f1 h1 i0 g0
e1 i1 j1 f1
d1 j1 k0 e0
c1 k1 l1 d1
b1 l1 a0 c0

a1 b1
 The story so far

Gethner: Are 2-blowups of planar graphs biplanar?

▶ Yes for triangle-free planar graphs [known]

▶ Yes for graphs that can be decomposed into two outerpaths [new]

▶ In general, they have thickness ≤ 3 [known]

Our main result

No!
 Our counterexamples

Kleetope of a polyhedron: glue a pyramid onto each face

Iterated Kleetope: do the same thing repeatedly, some number of times

Why the outerpath decomposition doesn’t work

When we glue in a pyramid, each dual path can

only visit two of the three new faces

k-level Kleetope ⇒ # faces expands as 3k

Longest path length expands only as 2k

Dual paths too short for two to cover all faces

 Why iterated Kleetopes can be hard to draw

The original graph is a subgraph of its Kleetope ⇒

Drawings of Kleetope contain drawings of original

If drawing a graph is difficult,

drawings its Kleetope can be more difficult

Iterating can make difficulties pile up enough to

make drawing become impossible
 Counting again
Blowups of maximal planar have only twelve fewer edges than max biplanar
⇒ In any drawing, most faces are triangles
⇒ ∃ vertex whose four images (×2 from blowup, ×2 from biplanar)
are surrounded entirely by triangular faces
 Piling on constraints

There exists a vertex whose four images . . .

▶ . . . are each surrounded by triangular faces (max planar)
▶ . . . are each surrounded by three triangles (Kleetope)
▶ . . . are each surrounded by three triangles, sharing ≤ 1
edge with triangles around other images (Kleetope2 )
▶ . . . are each surrounded by three triangles, edge-disjoint
with triangles around other images (Kleetope3 )
⇒ impossible to draw
 Generalization: split thickness
Allow each country to have a colony on
the Earth, not on the moon
⇒ drawing with two copies of each node,
in a single plane (here: K6,10 )
[Heawood 1890; Eppstein et al. 2018]

∃ drawings of 2-blowups of more graphs

E.g.: Kleetopes of outerpath
decompositions

Same proof shows that 2-blowups of

iterated Kleetopes of large maximal planar
graphs do not have split thickness two
 Summary

Some planar graphs (iterated Kleetopes of large

polyhedra) have non-biplanar 2-blowups

Same proof shows that 2-blowup does not have split

thickness two

However, graphs with two-outerpath decompositions

have biplanar blowups
Their Kleetopes have split thickness two blowups
 Some open problems

Are 2-blowups of 3-colorable planar graphs biplanar?

(Kleetopes are not 3-colorable)

Are 2-blowups of 4-vertex-connected planar graphs biplanar?

(Kleetopes are not 4-vertex-connected)

Biplanarity is NP-complete [Mansfield 1983]; is it hard on 2-blowups of planar graphs?

How hard is it to find our two-outerpath decompositions?

Can our two-outerpath drawing be strengthened to geometric thickness?

(Straight-line plane drawing with two edge colors, no monochromatic crossings)
 References and image credits, I

Michael O. Albertson, Debra L. Boutin, and Ellen Gethner. The thickness and
chromatic number of r -inflated graphs. Discrete Mathematics, 310(20):2725–2734,
2010. doi: 10.1016/j.disc.2010.04.019.
Alan Bennett. Klein bottle. CC-BY-NC-SA licensed image, Science Museum Group,
1995. URL https://collection.sciencemuseumgroup.org.uk/objects/
co415792/klein-bottle-1995-single-surface-model.
David Eppstein, Philipp Kindermann, Stephen Kobourov, Giuseppe Liotta, Anna Lubiw,
Aude Maignan, Debajyoti Mondal, Hamideh Vosoughpour, Sue Whitesides, and
Stephen Wismath. On the planar split thickness of graphs. Algorithmica, 80(3):
977–994, 2018. doi: 10.1007/s00453-017-0328-y.
Ellen Gethner. To the Moon and beyond. In Ralucca Gera, Teresa W. Haynes, and
Stephen T. Hedetniemi, editors, Graph Theory: Favorite Conjectures and Open
Problems, II, Problem Books in Mathematics, pages 115–133. Springer International
Publishing, 2018. doi: 10.1007/978-3-319-97686-0_11.
 References and image credits, II

P. J. Heawood. Map colour theorem. Quarterly Journal of Mathematics, 24:332–338,

1890.
Anthony Mansfield. Determining the thickness of graphs is NP-hard. Mathematical
Proceedings of the Cambridge Philosophical Society, 93(1):9–23, 1983. doi:
10.1017/S030500410006028X.
NASA. The Earth straddling the limb of the Moon, as seen from above Compton
crater. Public domain image, October 12 2015. URL
https://commons.wikimedia.org/wiki/File:
EarthFromTheMoon-LRO-20151012a.jpg.
C. St. J. A. Nash-Williams. Decomposition of finite graphs into forests. J. London
Math. Soc., 39:12, 1964. doi: 10.1112/jlms/s1-39.1.12.
Pat Rawlings. Lunar mining facility. Public domain image, 1995. URL
https://commons.wikimedia.org/wiki/File:Mooncolony.jpg.
 References and image credits, III

Gerhard Ringel. Färbungsprobleme auf Flächen und Graphen, volume 2 of

Mathematische Monographien. VEB Deutscher Verlag der Wissenschaften, Berlin,
1959.
Sesame Street. Number of the Day Waltz: 12. YouTube video, 2005. URL
https://www.youtube.com/watch?v=GAUSzWo0FyM.
Unknown. Ridiculous parking sign in Culver City, California. Unattributed web image,
undated. URL https://www.reddit.com/r/LosAngeles/comments/p6eao0/
does_anyone_know_where_this_ridiculous_parking/.