Mazur’s Growth Number Conjecture in the Rank One Case

Debanjana Kundu, Antonio Lei · 2025

Read Mazur’s Growth Number Conjecture in the Rank One Case by Debanjana Kundu, Antonio Lei (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-2504-10761.txt.

Introduction

Debanjana Kundu and Antonio Lei prove a case of a growth conjecture in a 2025 math paper. The conjecture comes from Barry Mazur and counts how fast a certain rank grows in a tower of number fields. The team works with an elliptic curve that behaves well at a prime of five or more. A special point on the curve, called a Heegner point, holds the key. When that point's height measure is not zero, the conjecture holds in the rank one case, they show.

Catalog record

CREATORDebanjana Kundu, Antonio Lei
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE38 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-10761

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Mazur’s Growth Number Conjecture in the Rank One Case Debanjana Kundu, Antonio Lei Abstract . Let p ≥ 5 p\geq 5 be a prime number. Let 𝖤 / ℚ \mathsf{E}/\mathbb{Q} be an elliptic curve with good ordinary reduction at p p . Let K K be an imaginary quadratic field where p p splits, and such that the generalized Heegner hypothesis holds. Under mild hypotheses, we show that if the p p -adic height of the Heegner point of 𝖤 \mathsf{E} over K K is non-zero, then Mazur’s conjecture on the growth of Selmer coranks in the ℤ p 2 \mathbb{Z}_{p}^{2} -extension of K K holds. 2020 Mathematics Subject Classification Primary 11R23, 11G05 1. Introduction Let K be a number field. Let \mathsf{E} be an elliptic curve defined over K and L/K be an algebraic extension. The Mordell–Weil theorem asserts that when L/K is a finite extension, the group \mathsf{E}(L) of L -rational points is a finitely generated \mathbb{Z} -module. In other words, the (algebraic) rank of \mathsf{E}(L) is finite. However, when L is an infinite algebraic extension of K , the situation is more intricate. For an elliptic curve \mathsf{E}/\mathbb{Q} and p a prime number, the works of K. Kato [ Kat04 ] and D. Rohrlich [ Roh88 ] imply that over the cyclotomic \mathbb{Z}_{p} -extension \mathbb{Q}_{\mathrm{cyc}} of \mathbb{Q} , the (algebraic) rank of \mathsf{E}(\mathbb{Q}_{\mathrm{cyc}}) is finite. However, when K/\mathbb{Q} is an imaginary quadratic field, and K_{\mathrm{ac}} is the anti-cyclotomic \mathbb{Z}_{p} -extension of K , it is possible that \mathsf{E}(K_{\mathrm{ac}}) has infinite rank; see [ Gre99 , (1.9) or (1.10)] . In [ Maz83 , § 18, p. 201] , B. Mazur formulated a conjecture predicting that K_{\mathrm{ac}} is the only \mathbb{Z}_{p} -extension of K over which the Mordell–Weil rank can be infinite. Mazur’s Growth Number Conjecture . Suppose \mathsf{E}/\mathbb{Q} is an elliptic curve, K is an imaginary quadratic field, and p is a prime of good ordinary reduction for \mathsf{E} . Let \mathcal{K}/K be a \mathbb{Z}_{p} -extension, and write \mathcal{K}_{n} to denote the unique subfield of degree p^{n} of K in \mathcal{K} . Further suppose that every prime of bad reduction of \mathsf{E} is finitely decomposed in \mathcal{K}/K . Then for n\gg 0 \operatorname{corank}_{\mathbb{Z}_{p}}\mathrm{Sel}_{p^{\infty}}(\mathsf{E}/\mathcal{K}_{n})=cp^{n}+O(1), where c=\begin{cases}0&\textrm{ if }\mathcal{K}\neq K_{{\mathrm{ac}}}\text{ or }(\mathsf{E},K)\textrm{ has `sign' }+1\\ 1&\textrm{ if }\mathcal{K}=K_{{\mathrm{ac}}},(\mathsf{E},K)\textrm{ has `sign' }-1\textrm{ and is `generic'}\\ 2&\textrm{ if }\mathcal{K}=K_{{\mathrm{ac}}},(\mathsf{E},K)\textrm{ has `sign' }-1\textrm{ and is `exceptional'}.\end{cases} A few remarks are in order regarding the conjecture. The integer c is called the growth number . The pair (\mathsf{E},K) is called ‘generic’ if \mathsf{E} has no CM, or the CM field of \mathsf{E} is different from K . On the other hand, the pair (\mathsf{E},K) is called ‘exceptional’ if \mathsf{E} has CM by (an order in) K . The sign of (\mathsf{E},K) is defined in [ Maz83 , § 6] ) to mean • The sign of the functional equation of L(\mathsf{E}/K,s) in the ‘generic’ case and • The sign of the functional equation of L(\varphi,s) in the ‘exceptional’ case, where \varphi is the Hecke character of K satisfying L(\mathsf{E}/\mathbb{Q},s)=L(\varphi,s). It is pertinent to note that the ‘sign’ of (\mathsf{E},K) is not the same as the sign of the functional equation of L(\mathsf{E}/K,s) in the ‘exceptional’ case. In particular, predicts that c=0 precisely when \mathcal{K}\neq K_{{\mathrm{ac}}} or when one of the following situations is true • (\mathsf{E},K) is ‘generic’ and L(\mathsf{E}/K,s) has sign +1, • (\mathsf{E},K) is ‘exceptional’ and L(\mathsf{E}/\mathbb{Q},s)=L(\varphi,s) has sign +1. is often formulated in terms of the growth of Mordell–Weil ranks. The two formulations are equivalent under the condition that the p -parts of the Tate–Shafarevich groups are finite. We assume that this is the case in the discussion below. We record some instances in which the conjecture has been proven. a) The exceptional case of is well-understood from the works of K. Rubin, D. Rohrlich, B. Gross–D. Zagier, and R. Greenberg (see for example [ Gre99 , Theorems 1.7 and 1.8] ). Let \mathsf{E}/\mathbb{Q} be an elliptic curve with complex multiplication by an imaginary quadratic field K . If p is any prime, the Mordell–Weil rank of \mathsf{E} remains bounded in every \mathbb{Z}_{p} -extension which is different from K_{{\mathrm{ac}}} . Over the anti-cyclotomic \mathbb{Z}_{p} -extension K_{{\mathrm{ac}}}/K , the Mordell–Weil rank of a CM elliptic curve is unbounded if p is a prime of good supersingular reduction. More explicitly, for n\gg 0 \operatorname{rank}_{\mathbb{Z}}\mathsf{E}(K_{{\mathrm{ac}},n})-\operatorname{rank}_{\mathbb{Z}}\mathsf{E}(K_{{\mathrm{ac}},n-1})=\epsilon_{n}\phi(p^{n}), where \phi(p^{n}) is the Euler totient function and \epsilon_{n}=0 or 2, depending on the parity of n . On the other hand, if p is a prime of good ordinary reduction then the boundedness of the Mordell–Weil rank of \mathsf{E} in K_{{\mathrm{ac}}}/K depends on the parity of the order of vanishing of the Hasse–Weil L -function at s=1 . In particular, if L(\mathsf{E}/\mathbb{Q},s) has an odd (resp. even) order 0 at s=1 , then the rank growth is unbounded (resp. bounded). In the case of unbounded rank growth, the rank increases by exactly 2\phi(p^{n}) at the n -th layer for n\gg 0 . b) More recently, there has been some partial progress made towards this conjecture for non-CM elliptic curves in [ GHKL24 , KMS23 ] . Let K be an imaginary quadratic field in which p is unramified and write K_{\infty} for the \mathbb{Z}_{p}^{2} -extension of K . When p is a prime of good ordinary reduction, it was proven in [ KMS23 ] that the number of \mathbb{Z}_{p} -extensions of K where the rank of \mathsf{E} does not stay bounded is at most \min_{H}\{\lambda_{H}\} , where H runs over subgroups of \operatorname{Gal}(K_{\infty}/K) such that K_{\infty}^{H} is an admissible 1 1 1 in the sense of Definition 1.1 of op. cit. \mathbb{Z}_{p} -extension of K satisfying the \mathfrak{M}_{H}(G) -property, and \lambda_{H} is the \lambda -invariant of the Pontryagin dual of the p -primary Selmer group of E over that \mathbb{Z}_{p} -extension. In particular, the authors proved that holds when the Mordell–Weil rank of \mathsf{E}(K) is zero or one, under some hypotheses (see in particular Theorems 9.3 and 9.4 in op. cit. ). On the other hand, the authors in [ GHKL24 ] proved Mazur’s conjecture under a semi-cyclity hypothesis, which can be verified via computations, and is different from the hypotheses imposed in [ KMS23 ] . c) When \mathsf{E}/\mathbb{Q} has good supersingular reduction at p , an analogue of Mazur’s Conjecture was formulated in [ LS20 ] . For partial progress towards this conjecture in the case of non-CM elliptic curves, we refer the reader to [ LS20 , HL20 ] . For elliptic curves of rank 0 over K and p\geq 3 a prime of supersingular reduction with a_{p}=0 such that both primes above p in K are completely ramified in K_{{\mathrm{ac}}}/K , this conjecture was resolved in [ GHKL24 ] . The main goal of this article is to provide an analytic proof of for elliptic curves that are of rank 1 over K when p is a prime of good ordinary reduction. More precisely, we prove the following theorem. Theorem A . Fix a prime p\geq 5 . Let \mathsf{E}/\mathbb{Q} be a non-CM elliptic curve with good ordinary reduction at p . Let K be an imaginary quadratic field where p splits, and such that ( (GHH) ), ( h-IMC ) and (

Read the complete text