On refined nonvanishing conjectures by Kurihara and Kolyvagin
Francesc Castella, Takamichi Sano · 2026
Read On refined nonvanishing conjectures by Kurihara and Kolyvagin by Francesc Castella, Takamichi Sano (2026) 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-2601-14504.txt.
Introduction
Francesc Castella and Takamichi Sano extend number theory results in this 2026 paper. Their topic is refined nonvanishing conjectures by Kurihara and Kolyvagin. Such conjectures predict that certain algebraic quantities tied to elliptic curves do not vanish. Past work by others set out results on these conjectures in limited cases.
The authors widen those results in two ways. For Kurihara's conjectures, they allow primes of any reduction type. For Kolyvagin's, they allow inert primes in the underlying imaginary quadratic field. Readers get stronger forms of two refined conjectures, covering prime cases that earlier proofs had to exclude.
Catalog record
| CREATOR | Francesc Castella, Takamichi Sano |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 76 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-14504
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On refined nonvanishing conjectures by Kurihara and Kolyvagin Francesc Castella, Takamichi Sano Abstract . In this paper, we extend the results of [ 5 ] on refined conjectures by Kurihara and Kolyvagin, allowing primes of any reduction type in the case of Kurihara’s conjectures, and inert primes in the underlying imaginary quadratic field in the case of Kolyvagin’s. The key innovation is a new approach to the computation of the p p -divisibility index of certain special elements in Galois cohomology (the bottom class of a Λ \Lambda -adic Euler system twisted by a character sufficiently close to the trivial character) based on a reformulation of the Iwasawa Main Conjectures in terms of determinants of Selmer complexes. 1. Introduction In [ 5 ] , the first author with Burungale, Grossi and Skinner gave a proof of nonvanishing conjectures by Kolyvagin [ 29 ] and Kurihara [ 32 ] , and their ‘quantitative’ refinements [ 51 , 26 ] , for elliptic curves E/\mathbb{Q} at certain primes p . Building on a new approach to the descent computations in [ 5 ] , in this paper we obtain an extension of the results in op. cit. allowing E to have any reduction type at p in the case of the refined Kurihara conjectures, and p to be any prime unramified in the underlying quadratic imaginary field K in the case of the refined Kolyvagin conjectures. 1.1. Main results 1.1.1. Kurihara’s analytic quantities Let E/\mathbb{Q} be an elliptic curve of conductor N without complex multiplication, and fix an odd prime p such that (sur) \textrm{$\bar{\rho}:G_{\mathbb{Q}}=\operatorname{Gal}(\overline{\mathbb{Q}}/\mathbb{Q})\to{\rm Aut}_{\mathbb{F}_{p}}(E[p])$ is surjective}. Denote by \mathcal{L} the set of primes \mathcal{L}:=\{\ell\;\colon\;\textrm{$\ell\equiv 1\;({\rm mod}\,p)$,\; $\ell\nmid N$,\; and\; $a_{\ell}\equiv\ell+1\;({\rm mod}\,p)$}\}, where a_{\ell}:=\ell+1-\#\widetilde{E}(\mathbb{F}_{\ell}) , and let \mathcal{N} be the collection of all squarefree products of primes \ell\in\mathcal{L} , with the convention that 1\in\mathcal{N} (corresponding to the empty product). For each \ell\in\mathcal{L} , let I_{\ell}\subset\mathbb{Z} be defined by I_{\ell}:=(\ell-1,a_{\ell}-\ell-1)\mathbb{Z}_{p}\cap\mathbb{Z}, and for each n\in\mathcal{N} put I_{n}=\sum_{\ell\mid n}I_{\ell} if n\neq 1 and I_{n}=0 otherwise. Let f\in S_{2}(\Gamma_{0}(N)) be the newform attached to E . Let \Omega_{E}^{+}=\int_{E(\mathbb{R})}\omega_{E}\in\mathbb{R} be the positive Néron period of E , where \omega_{E} is a Néron differential. Fix a modular parametrization \phi:X_{0}(N)\rightarrow E and let c_{\phi}\in\mathbb{Z} be the associated Manin constant , so that \phi^{*}(\omega_{E})=c_{\phi}\cdot 2\pi if(z)dz . Assume that ( p\nmid c_{\phi} ) p\nmid c_{\phi}. (For instance, if E is the strong Weil curve in the \mathbb{Q} -isogeny class attached to f and \phi is the optimal quotient, then ( ∤ p c ϕ ) holds provided p^{2}\nmid N by [ 35 , Cor. 4.1] .) Using modular symbols, Kurihara introduced in [ 31 ] certain quantities \delta_{n}\in\mathbb{Z}/I_{n} shown to completely determine the structure of the p -primary Selmer group {\rm Sel}_{p^{\infty}}(E/\mathbb{Q}) under some hypotheses (see [ 31 , Thm. B] ). Specifically, \delta_{n}:=\sum_{\begin{subarray}{c}a=1\\ (a,n)=1\end{subarray}}^{n}\overline{\left[\frac{a}{n}\right]}\Biggl(\prod_{\ell\mid n}{\rm log}_{\mathbb{F}_{\ell}}(a)\Biggr)\in\mathbb{Z}/I_{n}, where \left[\frac{a}{n}\right]\in\mathbb{Q} is the modular symbol \left[\frac{a}{n}\right]={\rm Re}\biggl(2\pi i\int_{\infty}^{a/n}f(z)dz\biggr)/\Omega_{E}^{+}\in\mathbb{Q}, which is known to be p -integral by our assumptions on p (see [ 46 , §3] ), and \overline{\left[\frac{a}{n}\right]}\in\mathbb{Z}/I_{n} denotes its reduction modulo I_{n} ; and {\rm log}_{\mathbb{F}_{\ell}}:\mathbb{F}_{\ell}^{\times}\cong\mathbb{Z}/(\ell-1)\mathbb{Z}\rightarrow\mathbb{Z}/I_{\ell} is the discrete logarithm defined by a fixed choice of primitive root \eta_{\ell}\in\mathbb{F}_{\ell}^{\times} (so \mathbb{F}_{\ell}^{\times}=\eta_{\ell}^{\mathbb{Z}} ), with the last arrow given by reduction modulo I_{\ell} . In particular, for n=1 , \delta_{n} reduces to the normalized L -value (1.1) \delta_{1}=L(E,1)/\Omega_{E}^{+}\in\mathbb{Q}. 1.1.2. Kurihara’s conjectures Put {\rm Tam}_{E}:=\prod_{\ell\mid N}c_{\ell}, where c_{\ell}=[E(\mathbb{Q}_{\ell}):E_{0}(\mathbb{Q}_{\ell})] is the Tamagawa factor of E at \ell . Let \bar{\delta}_{n}\in\mathbb{F}_{p} denote the reduction of \delta_{n} modulo p . The following conjecture was first formulated by Kurihara in [ 32 , §1.2, Conj. 1] in the p -ordinary case, and in [ 33 , §1.2, Conj. 1.1] for supersingular p . As in [ 26 ] , it naturally extends to primes p>2 of bad reduction as long as hypotheses ( sur ) and ( ∤ p c ϕ ) both hold, and below we state it in this level of generality. Conjecture A . Let p be an odd prime such that ( sur ) and ( ∤ p c ϕ ) both hold. If p\nmid{\rm Tam}_{E} , then there exists n\in\mathcal{N} such that \bar{\delta}_{n}\neq 0 . 1.1.3. Refined Kurihara’s conjecture In this paper we study a strenghtening of Conjecture A motivated by the aforementioned structure theorem by Kurihara [ 31 ] . The same structure theorem was recently proved from a different perspective by C.-H. Kim [ 26 ] , whose notations we largely follow. For each n\in\mathcal{N} define \mathscr{M}(n)\in\mathbb{Z}_{\geq 0}\cup\{\infty\} by \mathscr{M}(n):=\infty if \delta_{n}=0 , and by \mathscr{M}(n):=\max\{\mathscr{M}\geq 0\;\colon\;\delta_{n}\in p^{\mathscr{M}}\mathbb{Z}/I_{n}\} otherwise. Put \mathscr{M}_{r}:=\min\{\mathscr{M}(n)\;\colon\;n\in\mathcal{N},\;\nu(n)=r\} , where \nu(n) denotes the number of prime factors of n . As shown in [ 31 , §10.5] , the quantity \delta_{n} can be obtained as a coefficient of the Mazur–Tate \theta -element of conductor n [ 34 ] , and it follows from the functional equation of the latter that \delta_{n}=0 unless (-1)^{\nu(n)}=\epsilon , where \nu(n) where \epsilon\in\{\pm{1}\} is the root number of E . Putting \mathscr{M}_{r}:=\min\{\mathscr{M}(n)\;\colon\;\nu(n)=r\} , then one can show that \mathscr{M}_{r}\geq\mathscr{M}_{r+2}\geq 0 for all r\geq 0 , and we put \mathscr{M}_{\infty}(\bm{\delta}):=\lim_{\begin{subarray}{c}r\to\infty\\ (-1)^{r}=\epsilon\end{subarray}}\mathscr{M}_{r}, where \bm{\delta} denotes the collection \{\delta_{n}\}_{n\in\mathcal{N}} . Thus \mathscr{M}_{\infty}(\bm{\delta})<\infty if and only if \bm{\delta}\neq\{0\} . Suppose this is the case, and let \varrho be the ‘order of vanishing’ of \bm{\delta} : \varrho:=\min\{r\geq 0\;\colon\;\textrm{$\delta_{n}\neq 0$\; and\; $\nu(n)=r$}\}. Then [ 26 , Thm. 1.8] , as originally proved by Kurihara [ 31 , Thm. B] under some hypotheses and from a different perspective, yields the following exact formula for \hbox{\russ\char 88\relax}_{\rm BK}(E[p^{\infty}]/\mathbb{Q}):={\rm Sel}_{p^{\infty}}(E/\mathbb{Q})/{\rm Sel}_{p^{\infty}}(E/\mathbb{Q})_{\rm div} , where the subscript ‘ {\rm div} ’ denotes the maximal divisible submodule: {\rm ord}_{p}\bigl(\#\hbox{\russ\char 88\relax}_{\rm BK}(E[p^{\infty}]/\mathbb{Q})\bigr)=\mathscr{M}_{\varrho}-\mathscr{M}_{\infty}(\bm{\delta}). In particular, if L(E,1)\neq 0 , then \varrho=0 and Kato’s work [ 24 ] implies that \hbox{\russ\char 88\relax}_{\rm BK}(E[p^{\infty}]/\mathbb{Q}) is the same as the p -primary part \hbox{\russ\char 88\relax}(E/\mathbb{Q})[p^{\infty}] of the Tate–Shafarevich group of E , so combined with ( 1.1 ) the above exact formula reduces to {\rm ord}_{p}\bigl(\#\hbox{\russ\char 88\relax}(E/\mathbb{Q})[p^{\infty}]\bigr)={\rm ord}_{p}\biggl(\frac{L(E,1)}{\Omega_{E}^{+}}\biggr)-\mathscr{M}_{\infty}(\bm{\delta}). Together with the Birch–Swinnerton-Dyer formula, this motivates the following conjecture (in arbitrary rank) formulated in [ 26 , Conj. 1.9] . Conjecture B (Refined Kurihara’s conjecture) . Let p>3 be a prime such that ( sur ) and ( ∤ p c ϕ ) both hold. Then \mathscr{M}_{\infty}(\bm{\delta})={\rm ord}_{p}({\rm Tam}_{E}). 1.1.4. Results Our main result towards the refined Kurihara’s conjecture is the following. Let \mathbb{Q}_{\infty}/\mathbb{Q} denote the cyclotomic \mathbb{Z}_{p} -extension. Theorem A . Let p>3 be a prime such that ( sur ) and ( ∤ p c ϕ ) both hold. Then the following are equivalent: (i) \mathscr{M}_{\infty}(\bm{\delta})={\rm ord}_{p}({\rm Tam}_{E}) , and hence Conjecture B holds. (ii) The Iwasawa Main Conjecture 2.2.2 for \mathbb{Q}_{\infty}/\mathbb{Q} holds. In particular, in the case where p\nmid{\rm Tam}_{E} we deduce: Corollary A . Let p>3 be a prime such that ( sur ) and ( ∤ p c ϕ ) both hold. If p\nmid{\rm Tam}_{E} , then the following are equivalent: (i) There exists n\in\mathcal{N} such that \bar{\delta}_{n}\neq 0 , and hence Conjecture A holds. (ii) The Iwasawa Main Conjecture 2.2.2 for \mathbb{Q}_{\infty}/\mathbb{Q} holds. Proof. This is clear from Theorem A , noting that \mathscr{M}_{\infty}(\bm{\delta})=0 if and only if there exists n\in\mathcal{N} such that \delta_{n} is p -indivisible. ∎ The Iwasawa Main Conjecture 2.2.2 for \mathbb{Q}_{\infty}/\mathbb{Q} in the above results is a reformulation of