\(\Lambda\)-adic Euler characteristics of elliptic curves (Q2369730)

Let \(E/{\mathbb Q}\) be a modular elliptic curve, and \(p > 3\) a good ordinary or semi-stable prime, \(\Gamma = 1 + p {\mathbb Z}_p = \langle u \rangle\) and \(\wedge = {\mathbb Z}_p [[ \Gamma]]\) the usual Iwasawa algebra. Let \(f_2\) be the \(p\)-stabilisation of the normalised eigenform \(f_E\) at \(p\). If \(f_2\) is the unique \(p\)-stabilised newform in \(S^{\text{ord}}_2 (\Gamma_0 (N p))\), there exists a unique \(\wedge\)-adic eigenform \(f \in \wedge [[q]]\) lifting \(f_2\) at weight 2; furthermore, \(f_k := \sum^\infty_{n=1} \sigma_k (a_n(f)) q^n\), where \(\sigma_k (u) = u^{k-2}\), is a \(p\)-stabilised eigenform of weight \(k\) and character \(\omega^{k-2}\), for any integer \(k \geq 2\). Hida and Mazur-Wiles have constructed a continuous Galois representation \(\rho_\infty: G_{\mathbb Q} \to \text{GL}_2 (\wedge)= \Aut_\wedge (T_\infty)\) interpolating Deligne's \(p\)-adic representations for every eigenform in the family. Here \(T_\infty\) is a rank 2 lattice over \(\wedge,\) unramified outside \(Np,\) cut out of the Tate module of the pro-Jacobian \(\varprojlim_r \operatorname{Jac} X_1 (N p^r)\) by \textit{H. Hida}'s method [Invent. Math. 85, 545--613 (1986; Zbl 0612.10021)]. For a finite set \(\Sigma\) containing \(p\) and the primes dividing the conductor of \(E,\) the author aims to study a ``big Selmer group'' \(\text{Sel}_{\mathbb Q} (\rho_\infty)\) which lives inside \(H^1 ({\mathbb Q}_\Sigma/{\mathbb Q}, \Hom_{\text{cont}}(T_\infty, \mu_{p^\infty}))\) and which interpolates the Bloch-Kato Selmer groups associated to the \(p\)-stabilisations \(f_k\) of weight \(k \geq 2.\) Let \(\text Ш_{\mathbb Q} (\rho_\infty)\) be the characteristic series of the \(\wedge\)-cotorsion of \(\text{Sel}_{\mathbb Q}(\rho_\infty).\) The author defines \(\wedge\)-adic Tamagawa numbers which control the specialisation of the ``big Tate-Shafarevich group'' at arithmetic points. In particular, for the weight \(k=2\), they occur in the leading term of \(\text Ш_{\mathbb Q} (\rho_\infty)\). Actually the author's main theorem is a formula giving the \(p\)-adic valuation of this leading term in the style of the Tamagawa number conjecture. As an example of application, he shows that the \(\mu\)-invariant \(\mu^{wt}\) of \(\text Ш_{\mathbb Q}(\rho_\infty)\) is trivial for \(E= X_0(11)\) and \(5<p<100.\) More generally, the behavior of \(\mu^{wt}\) and of the \(\wedge\)-adic Tamagawa numbers are related to the arithmetic of \(F_\infty= F(E[p^\infty])\) as follows: suppose for simplicity that \(E\) has no complex multiplication; under the twin assumptions that \(L(E, 1)\neq 0\) and \(\text{Sel}_{F_\infty}(E)\) is cotorsion over the non-abelian Iwasawa algebra, \textit{J. Coates} and \textit{S. Howson} [J. Math. Soc. Japan 53, 175--235 (2001; Zbl 1046.11079)] have given for the Euler-Poincaré characteristic \(\chi (\text{Gal} (F_\infty/{\mathbb Q}), \text{Sel}_{F_\infty}(E))\) an expression in the style of the Birch and Swinnerton-Dyer conjecture. Let \(\mu^{\text{GL}_2}\) be the \(p\)-adic valuation of this Euler-Poincaré characteristic. Then the author's main theorem yields the upper bound: \[ \mu^{wt} \leq \mu^{\text{GL}_2} + \sum_{\text{bad primes}} \{\text{ord}_p (L_\ell (E, 1))- \text{ord}_p (\text{Tam}_{\wedge, \ell})\}. \] In other words, the arithmetic of the weight-deformation is controlled in the \(\text{GL}_2\)-extension, in accordance with what is expected from a ``non commutative main conjecture'' in Iwasawa theory.