Ceresa cycles of X 0 ( N ) X_{0}(N)
Elvira Lupoian and James Rawson · 2025
Read Ceresa cycles of X 0 ( N ) X_{0}(N) by Elvira Lupoian and James Rawson (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-2501-14060.txt.
Introduction
Elvira Lupoian and James Rawson study in 2025 a subtle cycle attached to modular curves, a classic family of curves tied to number theory. The Ceresa cycle is invisible to homology, the coarse shape test, yet Ceresa showed it is usually not zero in the finer Chow group. The pair ask when the cycle is merely torsion for the modular curve family.
For prime levels where the curve is not hyperelliptic, they prove the cycle has infinite order. For general levels, only finitely many curves can have a torsion cycle. Their route runs through special points on the curve's Jacobian, shown to have infinite order by the geometry of modular Jacobians.
Catalog record
| CREATOR | Elvira Lupoian and James Rawson |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-14060
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Ceresa cycles of X 0 ( N ) X_{0}(N) Elvira Lupoian and James Rawson Abstract . The Ceresa cycle is an algebraic 1-cycle on the Jacobian of an algebraic curve. Although it is homologically trivial, Ceresa famously proved that for a very general complex curve of genus at least 3, it is non-trivial in the Griffiths group. In this paper we study the Ceresa cycle attached to the complete modular curve X 0 ( N ) X_{0}(N) modulo rational equivalence. For prime level p p we give a complete description, namely we prove that if X 0 ( p ) X_{0}(p) is not hyperelliptic, then its Ceresa cycle is non-torsion. For general level N N , we prove that there are finitely many X 0 ( N ) X_{0}(N) with torsion Ceresa cycle. Our method relies on the relationship between the vanishing of the Ceresa cycle and Chow-Heegner points on the Jacobian. We use the geometry and arithmetic of modular Jacobians to prove that such points are of infinite order and therefore deduce non-vanishing of the Ceresa cycle. 2000 Mathematics Subject Classification 14C25, 14G35, 11G18 1. Introduction Let C be a smooth, projective and geometrically integral curve, defined over an algebraically closed field K of characteristic 0 . Fix a degree 1 base point divisor e and let \iota_{e} be the corresponding Abel-Jacobi map, which embeds the curve into its Jacobian J . The Ceresa cycle associated to (C,e) is defined as \Cer(C,e):=[\iota_{e}(C)]-[-1]^{*}[\iota_{e}(C)] which is an element of the Chow group \CH_{1}(J) , modulo rational equivalence. This cycle is always homologically trivial, that is, it lies in the kernel of the cycle class map from \CH_{1}(J) into a Weil cohomology theory, but crucially it is not always trivial in the Chow group. In fact, for a very general complex curve of genus at least 3, Ceresa [ 10 ] proved that it is algebraically, and in particular rationally, non-trivial independent of the choice of base point divisor. In the case of hyperelliptic curves, taking e to be a Weierstrass point shows that \Cer(C,e)=0 . Beyond this, there are few other explicit examples of curves where the vanishing/non-vanishing of the Ceresa cycle is known. 1.1. Main Results In this paper, we study a classically interesting family of curves, namely the modular curves whose non-cuspidal points parameterise isomorphism classes of elliptic curves with a marked isogeny. We study the Ceresa cycle of X_{0}(N) inside \CH_{1}(J_{0}(N))\otimes\mathbb{Q} , henceforth simply \CH_{1}(J_{0}(N)) . As detailed in Theorem 2.1 (cf. [ 50 ] ), the Ceresa cycle can only possibly vanish for one choice of base point divisor class, namely e=\frac{K_{C}}{2g-2} , where K_{C} is the canonical class and g is the genus of the curve. For the rest of this paper we write \Cer(N) for \Cer\left(X_{0}(N),\frac{K_{X_{0}(N)}}{2g-2}\right) . The first main result of our paper gives a complete classification for prime level X_{0}(N) . Theorem 1.1 . Let p be a prime number. The Ceresa cycle \Cer(p) is non-zero if and only if X_{0}(p) is not hyperelliptic. That is, if p>71 or p\in\{43,53,61,67\} , then \Cer(p) is non-trivial. As \Cer(p) is zero whenever X_{0}(p) is hyperelliptic, Theorem 1.1 is optimal. The vanishing of the Ceresa cycle on X_{0}(N) is related to certain (scaled) Chow-Heegner divisors, first introduced by Darmon, Rotger and Sols [ 13 ] . More specifically, for any correspondence on X_{0}(N) , \alpha , we define a rational divisor on the curve \Sh(\alpha) , which is referred to as a shadow point or shadow divisor throughout this paper. It follows (Proposition 2.2 ) that if \Cer(N) vanishes, then \Sh(\beta) defines a rational torsion point on the Jacobian J_{0}(N) for any correspondence \beta on X_{0}(N) , and this gives us a strategy for the proof of Theorem 1.1 . Hecke operators provide a natural source of correspondences on X_{0}(N) , and we proceed by showing that for certain Hecke operators T , the linear equivalence class, D , of the divisor \Sh(T) is non-torsion in J_{0}(p)(\mathbb{Q}) , systematically, depending on the congruence class of p modulo 12 . For p\equiv 5,7\mod{12} , we set D to be the linear equivalence class of 3\Sh(T_{2}) or 6\Sh(T_{2}) , depending on the congruence classes of p modulo 8 and 7 . We compute these divisors explicitly, in terms of CM points and cusps, using the definition of T_{2} and the moduli interpretation of the curve. We prove that D is non-torsion by projecting it on the Jacobian of X_{0}^{+}(p) , the quotient of X_{0}(p) by the Fricke involution w_{p} . The Jacobian of this curve has no non-zero rational torsion, and we proceed by reducing D modulo p and by arguing that this divisor is not cut out by a function on the curve using the special fibre of X_{0}^{+}(p) at p and explicit j -invariants, see § 5.2 for details. A similar approach is used for p\equiv 1\mod{12} , unless p\equiv 1\mod{8} and p\equiv 1,2,4\mod{7} , in which case we show that either 6\Sh(T_{2}) or 6\Sh(T_{3}) is non-torsion by studying the relations on j -invariants that would be imposed if both were zero. See § 5.3 for details. In the case of p\equiv 11\mod{12} , we cannot guarantee that \Sh(T_{\ell})\neq 0 for any fixed \ell . Instead, we work \Sh(w_{p}) , where w_{p} is the Fricke involution on X_{0}(p) , and we apply Proposition 3.4 (see [ 39 ] ), which controls the degrees of certain functions on X_{0}^{+}(p) , to deduce that for all p\geq 104,711 , \Sh(w_{p}) is non-torsion in the Jacobian. We deal with the remaining cases by finding an explicit \ell for which \Sh(T_{\ell})\neq 0 , and we apply our previous reduction method to this point. Our proof makes explicit use of the arithmetic of the J_{0}(p) , its rational points J_{0}(p)(\mathbb{Q}) and of the ground breaking work of Mazur [ 39 ] . Although the divisors studied in this work are supported on CM points on the curve, they are not ordinary Heegner points, and one cannot simply apply the Gross-Zagier theorem [ 27 ] , to conclude that they are non-torsion. To the best of our knowledge, this is the first work in which reduction (at bad primes) and related arithmetic tricks have been used to prove certain points on J_{0}(p)(\mathbb{Q}) are systematically non-torsion. Our second main result deals with the case of arbitrary level N and has significantly more exclusions. Theorem 1.2 . There are finitely many N such that \Cer(N)=0 . More precisely, if N>2^{5}\times 3^{4}\times 5^{2}\times 7^{2}\times\displaystyle\prod_{\begin{subarray}{c}11\leq p\leq 71,\\ p\not\in\{43,53,61,67\}\end{subarray}}p then \Cer(N) is non-zero in \CH_{1}(J_{0}(N)) . Combining Proposition 6.1 (which describes how Ceresa cycles behave in covers) and Theorem 1.1 , the proof of Theorem 1.2 , reduces to showing that for any prime p\in\{2,3,5,7,11,13,17,19,23,29,31,37,41,47,59,71\} \Cer(p^{n}) is non-torsion for some n\geq 2 . We deal with these finitely many computations explicitly. More precisely, we compute explicit shadow divisors, and show that they are non-torsion by applying the Gross-Zagier formula in some special cases or by explicit period lattice computations. These computations are explained in § 6 . While it is conceivably possible to use the methods of Section § 6 to treat all levels N for which X_{0}(N) is non-hyperelliptic and J_{0}(N)(\mathbb{Q}) has