Unicellular Maps Research Articles

Counting unicellular maps on non-orientable surfaces

A unicellular map is the embedding of a connected graph in a surface in such a way that the complement of the graph is a topological disk. In this paper we present a bijective link between unicellular maps on a non-orientable surface and unicellular maps of a lower topological type, with distinguished vertices. From that we obtain a recurrence equation that leads to (new) explicit counting formulas for non-orientable unicellular maps of fixed topology. In particular, we give exact formulas for the precubic case (all vertices of degree 1 or 3), and asymptotic formulas for the general case, when the number of edges goes to infinity. Our strategy is inspired by recent results obtained by the second author for the orientable case, but significant novelties are introduced: in particular we construct an involution which, in some sense, “averages” the effects of non-orientability.

Counting unicellular maps on non-orientable surfaces

A unicellular map is the embedding of a connected graph in a surface in such a way that the complement of the graph is a topological disk. In this paper we give a bijective operation that relates unicellular maps on a non-orientable surface to unicellular maps of a lower topological type, with distinguished vertices. From that we obtain a recurrence equation that leads to (new) explicit counting formulas for non-orientable precubic (all vertices of degree 1 or 3) unicellular maps of fixed topology. We also determine asymptotic formulas for the number of all unicellular maps of fixed topology, when the number of edges goes to infinity. Our strategy is inspired by recent results obtained for the orientable case [Chapuy, PTRF 2010], but significant novelties are introduced: in particular we construct an involution which, in some sense, ``averages'' the effects of non-orientability. \par Une carte unicellulaire est le plongement d'un graphe connexe dans une surface, tel que le complémentaire du graphe est un disque topologique. On décrit une opération bijective qui relie les cartes unicellulaires sur une surface non-orientable aux cartes unicellulaires de type topologique inférieur, avec des sommets marqués. On en déduit une récurrence qui conduit à de (nouvelles) formules closes d'énumération pour les cartes unicellulaires précubiques (sommets de degré 1 ou 3) de topologie fixée. On obtient aussi des formules asymptotiques pour le nombre total de cartes unicellulaires de topologie fixée, quand le nombre d'arêtes tend vers l'infini. Notre stratégie est motivée par de récents résultats dans le cas orientable [Chapuy, PTRF, 2010], mais d'importantes nouveautés sont introduites: en particulier, on construit une involution qui, en un certain sens, "moyenne'' les effets de la non-orientabilité.

The structure of unicellular maps, and a connection between maps of positive genus and planar labelled trees

A unicellular map is a map which has only one face. We give a bijection between a dominant subset of rooted unicellular maps of given genus and a set of rooted plane trees with distinguished vertices. The bijection applies as well to the case of labelled unicellular maps, which are related to all rooted maps by Marcus and Schaeffer’s bijection. This gives an immediate derivation of the asymptotic number of unicellular maps of given genus, and a simple bijective proof of a formula of Lehman and Walsh on the number of triangulations with one vertex. From the labelled case, we deduce an expression of the asymptotic number of maps of genus g with n edges involving the ISE random measure, and an explicit characterization of the limiting profile and radius of random bipartite quadrangulations of genus g in terms of the ISE.

A new combinatorial identity for unicellular maps, via a direct bijective approach.

We give a bijective operation that relates unicellular maps of given genus to unicellular maps of lower genus, with distinguished vertices. This gives a new combinatorial identity relating the number $\epsilon_g(n)$ of unicellular maps of size $n$ and genus $g$ to the numbers $\epsilon _j(n)$'s, for $j \lt g$. In particular for each $g$ this enables to compute the closed-form formula for $\epsilon_g(n)$ much more easily than with other known identities, like the Harer-Zagier formula. From the combinatorial point of view, we give an explanation to the fact that $\epsilon_g(n)=R_g(n) \mathrm{Cat}(n)$, where $\mathrm{Cat}(n$) is the $n$-th Catalan number and $R_g$ is a polynomial of degree $3g$, with explicit interpretation. On décrit une opération bijective qui relie les cartes à une face de genre donné à des cartes à une face de genre inférieur, portant des sommets marqués. Cela conduit à une nouvelle identité combinatoire reliant le nombre $\epsilon_g(n)$ de cartes à une face de taille $n$ et genre $g$ aux nombres $\epsilon _j(n)$, pour $j \lt g$. En particulier, pour tout $g$, cela permet de calculer la formule close donnant $\epsilon_g(n)$ bien plus facilement qu'à l'aide des autres identités connues, comme la formule d'Harer-Zagier. Du point de vue combinatoire, nous donnons une explication au fait que $\epsilon _g(n)=R_g(n) \mathrm{Cat}(n)$, où $\mathrm{Cat}(n)$ est le $n$ième nombre de Catalan et $R_g$ est un polynôme de degré $3g$, à l'interprétation explicite.

A bijection for covered maps on orientable surfaces

Abstract Unicellular maps are a natural generalisation of plane trees to higher genus surfaces. In this article we study covered maps, which are maps together with a distinguished unicellular spanning submap. We prove that the covered maps of genus g with n edges are in bijection with pairs made of a plane tree with n edges and a bipartite unicellular map of genus g with n + 1 edges. This generalises to any genus the bijection given in [O. Bernardi. Bijective counting of tree-rooted maps and shuffles of parenthesis systems. Electron. J. Combin., 14(1): Research Paper 9, 36 pp., 2007] between planar tree-rooted maps (maps with a distinguished spanning tree) and pairs made of a tree with n edges and a tree with n + 1 edges.

An analog of the Harer-Zagier formula for unicellular bicolored maps

An analog of the Harer-Zagier formula for unicellular bicolored maps