A Picard rank bound for base surfaces of elliptic Calabi-Yau 3-folds
2025
Read A Picard rank bound for base surfaces of elliptic Calabi-Yau 3-folds (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-06317.txt.
Introduction
Caucher Birkar and Seung-Joo Lee present a 2025 math paper that bounds a key count for surfaces used in building elliptic Calabi-Yau threefolds. These three-dimensional shapes matter in geometry and in string theory, where they set the stage for physical models. The count in question is the Picard rank, which measures how many independent divisor classes a surface carries. The authors compute an explicit upper bound for the rank of any compact surface that can serve as the base of such a threefold with mild singularities.
To get the bound, they develop a new strategy in birational geometry, partly inspired by six-dimensional physics models. The paper shows the derivation step by step, which also clarifies where the boundedness comes from. A companion paper covers the physics side.
Catalog record
| YEAR | 2025 |
|---|---|
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-06317
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A Picard rank bound for base surfaces of elliptic Calabi-Yau 3-folds Abstract We compute an explicit rank bound on the Picard group of the compact surfaces, which can serve as the base of an elliptic Calabi-Yau variety with canonical singularities. To bound the Picard rank from above, we develop a novel strategy in birational geometry, motivated in part by the physics of six-dimensional vacua of F-theory as discussed in a companion paper [ 1 ] to this one. The derivation of the concrete bound illustrates the strategy and clarifies the origin of the boundedness. Caucher Birkar a, 1 1 1 [email protected] and Seung-Joo Lee b, 2 2 2 [email protected] a Yau Mathematical Sciences Center, Jingzhai, Tsinghua University, Hai Dian District, Bejing, China 100084 b Department of Physics, Yonsei University, Seoul 03722, Republic of Korea 1 Introduction Addressing boundedness is a major goal to achieve in algebraic geometry. A variety of ideas and techniques have been developed to successfully establish boundedness of fibrations [ 2 ] , albeit mostly in an abstract setting. Independently in theoretical high energy physics, boundedness of the set of quantum gravity theories was suggested in the context of the swampland program [ 3 ] (see also [ 4 , 5 ] ). Since a plethora of effective quantum gravity theories arise from geometric compactifications of string theory, compact varieties allowed by string theory, notably, Calabi-Yau varieties, are expected to form a bounded family. While boundedness of Calabi-Yau varieties has remained an open issue in complex dimension d\geq 3 , that of elliptic Calabi-Yau varieties was successfully established in dimension d=3 [ 6 , 7 ] , and more recently, progress has been made in higher dimensions [ 8 ] . Elliptic Calabi-Yau varieties play a distinguished role also in string theory. In particular, in F-theory [ 9 ] , the base of an elliptic fibration is considered an internal space of Type IIB string theory and the fibration econdes the internal profile of the axio-dilation field, thereby fully geometrizing a string vacuum. Accordingly, boundedness of elliptic Calabi-Yau 3-folds amounts to that of the consistent six-dimensional N=(1,0) supergravity theories obtained as F-theory vacua. For applications in physics, it is then important to proceed further to promote the abstract boundedness notion to an explicit one. Notably, explicit upper-bounds on the spectrum of supergravity multiplets, if calculable, would inspire quantum field theorists and model builders alike. The aim of this paper is to present an explicit rank bound on the Picard group of the compact surface, which can serve as the base of an elliptic Calabi-Yau 3-fold. Our main theorem is the following. Theorem 1 Assume that • (Z,B) is a projective \frac{1}{6} -lc pair of dimension 2, • Z is a rational surface, • 12(K_{Z}+B)\sim 0 , then the Picard number \rho(Z)\leq 568 . The bound on the base Picard rank \rho in Theorem 1 immediately turns into an explicit bound on the tensor multiplet count, T , of the six-dimensional N=(1,0) vacua of F-theory: \rho\leq 568\quad\Leftrightarrow\quad T\leq 567\,. (1.1) Moreover, the algebro-geometric strategy we will develop in this paper for the bounding task has a natural counterpart in string theory, and more generally in effective quantum field theory. In this paper, however, we will focus exclusively on the formulations in algebraic geometry. Those readers interested in this connection between algebraic geometry and physics are kindly referred to a companion paper [ 1 ] to this one for more details. The plan of this paper is as follows. In the next section, for the purpose of establishing a Picard rank bound, we introduce a sequence of blowups and graphically visualize it with a tree. Building upon the preliminary results on the blowup sequence, in section 3 we systematically bound the number of blowups involved to prove our main theorem. Authors’ note: Having announced part of our results [ 10 ] at the stage of completing the work, we learned of the highly relevant paper [ 11 ] . It is worth pointing out that the explicit Picard rank bound of 568, as proposed by Theorem 1 , is by no means optimal. After all, as will be clear from the proof of the theorem, the derivation of this bound leaves much room for improvement and the concrete number presented should therefore be viewed in part as a proof of principle. It was conjectured [ 12 ] that the maximal Picard rank is 194 , which had been realized by a concrete F-theory vacuum [ 13 , 14 ] . In [ 11 ] this stronger bound is proven physically via the anomaly constraints, where the authors assume that the classification of superconformal theories and little string theories in [ 15 , 16 , 17 , 18 ] is complete and that light strings emerge in the boundary of the tensor branch [ 19 ] . It is thus our belief that we may use the bounding strategy developed in this paper in a more optimal fashion to push the bound down to 194 . 2 Preliminaries Lemma 2 Assume that • X is a 3-fold with canonical singularities • X\xrightarrow{h}Z is an elliptic fibration with K_{X}\equiv 0 over Z and with a rational section • K_{X}\equiv h^{*}(K_{Z}+\Lambda_{Z}+M_{Z}) is given by the canonical bundle formula where \Lambda_{Z} and M_{Z} are the discriminant and the moduli divisors, respectively. Then the pair (Z,\Lambda_{Z}+M_{Z}) is \frac{1}{6} -lc. Proof In general, a pair (V,B) is \epsilon -lc if for any resolution \varphi:W\to V , K_{W}+B_{W}=\varphi^{*}(K_{V}+B)\,, (2.1) the coefficients of B_{W} are less than or equal to 1-\epsilon . It is well-known that to show (Z,\Lambda_{Z}+M_{Z}) is \frac{1}{6} -lc, we can take an appropriate resolution of Z and take hyperplane sections to reduce the problem to the case with {\rm dim}(X)=2 [ 20 ] . So we can assume from now on that X\xrightarrow{h}Z is an elliptic fibration from a surface onto a curve, with K_{X}\equiv 0 over Z and h equipped with a section. Taking a resolution of X we can assume it is smooth. To show that the coefficients of \Lambda_{Z} at a point z\in Z is less than or equal to \frac{5}{6} is equivalent to showing that (X,\frac{1}{6}h^{*}z) is lc. 1 1 1 lc means 0 -lc. Here, the latter can be checked via the Kodaira classification; see the proof of Lemma 3 for more details. \Box Lemma 3 Under the assumption of Lemma 2 , the coefficients of \Lambda_{Z} belong to \{0,\frac{1}{6},\frac{1}{4},\frac{1}{3},\frac{1}{2},\frac{2}{3},\frac{3}{4},\frac{5}{6}\}\,. (2.2) Proof As in the proof of Lemma 2 , we can reduce to the case {\rm dim}(X)=2 . The coefficient of \Lambda_{Z} at a point z\in Z is equal to 1-t , where t={\rm sup}\{s\;|\;(X,s\,h^{*}z)\penalty\ \text{is}\penalty\ {\rm lc}\} is the lc-threshold. The possible fibers of h^{*}z are given on page 5 of [ 21 ] . We list in Table 2.1 the possible fiber types, as well as the corresponding t and the coefficient of \Lambda_{Z} computed for each fiber type. \Box Kodaira type lc threshold t coefficient of \Lambda_{Z} I_{0} 1 0 I_{1} 1 0 I_{n\geq 2} 1 0 I_{n\geq 0}^{*} \frac{1}{2} \frac{1}{2} II \frac{5}{6} \frac{1}{6} III \frac{3}{4} \frac{1}{4} IV \frac{2}{3} \frac{1}{3} IV^{*} \frac{1}{3} \frac{2}{3} III^{*} \frac{1}{4} \frac{3}{4} II^{*} \frac{1}{6} \frac{5}{6} Table 2.1: The lc threshold