Level $ 767$

Consider the modular Jacobian $ J_0(767)$ . Using the modular symbols package in Magma, one decomposes $ J_0(767)$ (up to isogeny) into a product of six optimal quotients of dimensions 2, 3, 4, 10, 17 and 23. The duals of these quotients are subvarieties $ {A_2}, {A_3}, {A_4},
A_{10}, {A_{17}}$ and $ {A_{23}}$ defined over $ \mathbb{Q}$ , where $ A_d$ has dimension $ d$ . Consider the subvariety $ A_{23}$ .

We first show that the Birch and Swinnerton-Dyer conjectural formula predicts that the orders of the groups $ {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A_{23})$ and $ {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A_{23}^\vee)$ are both divisible by 9.

Proposition 6.1.1   Assume [AS05, Conj. 2.2]. Then

$\displaystyle 3^2 \mid \char93 {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fon...
...wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A^{\vee}_{23}).
$

Proof. Let $ A = A^\vee_{23}$ . We use [AS05, §3.5 and §3.6] (see also [Ka81]) to compute a multiple of the order of the torsion subgroup $ A(\mathbb{Q})_{{\mathrm{tor}}}$ . This multiple is obtained by injecting the torsion subgroup into the group of $ \mathbb{F}_p$ -rational points on the reduction of $ A$ for odd primes $ p$ of good reduction and then computing the order of that group. Hence, the multiple is an isogeny invariant, so one gets the same multiple for $ A^\vee(\mathbb{Q})_{{\mathrm{tor}}}$ . For producing a divisor of $ \char93 A(\mathbb{Q})_{{\mathrm{tor}}}$ , we use the injection of the subgroup of rational cuspidal divisor classes of degree 0 into $ A(\mathbb{Q})_{{\mathrm{tor}}}$ . Using the implementation in Magma we obtain $ 120 \mid \char93 A(\mathbb{Q})_{{\mathrm{tor}}} \mid 240$ . To compute a divisor of $ A^\vee(\mathbb{Q})_{{\mathrm{tor}}}$ , we use the algorithm described in [AS05, §3.3] to find that the modular degree $ m_A = 2^{34}$ , which is not divisible by any odd primes, hence $ 15 \mid \char93 A^\vee(\mathbb{Q})_{{\mathrm{tor}}} \mid 240$ .

Next, we use [AS05, §4] to compute the ratio of the special value of the $ L$ -function of $ A_{/\mathbb{Q}}$ at 1 over the real Néron period $ \Omega_{A}$ . We obtain
$ \displaystyle \frac{L(A_{/\mathbb{Q}}, 1)}{\Omega_{A}} =
c_{A} \cdot \frac{2^9 \cdot 3}{5}$ , where $ c_{A}\in\mathbb{Z}$ is the Manin constant. Since $ c_{A} \mid 2^{\textrm{dim}(A)}$ by [ARS06] then

$\displaystyle \frac{L(A_{/\mathbb{Q}}, 1)}{\Omega_{A}} = \frac{2^{n+2} \cdot 3}{5},
$

for some $ 0 \leq n \leq 23$ . In particular, the modular abelian variety $ A_{/\mathbb{Q}}$ has rank zero over $ \mathbb{Q}$ .

Next, using the algorithms from [CS01,KS00] we compute the Tamagawa number $ c_{A,13} = 1920 = 2^3 \cdot 3 \cdot 5$ . We also find that $ 2 \mid c_{A,59}$ is a power of $ 2$ because $ W_{59}$ acts as $ 1$ on $ A$ , and on the component group $ {\mathrm{Frob}}_{59}=-W_{59}$ , so the fixed subgroup $ \Phi_{A,59}(\mathbb{F}_{59})$ of Frobenius is a $ 2$ -group (for more details, see [Rib90a, Prop. 3.7-8]).

Finally, the Birch and Swinnerton-Dyer conjectural formula for abelian varieties of Mordell-Weil rank zero (see [AS05, Conj. 2.2]) asserts that

$\displaystyle \frac{L(A_{/\mathbb{Q}},1)}{\Omega_{A}} = \frac{\char93 {\mbox{{\...
...hbb{Q})_{{\mathrm{tor}}} \cdot \char93  A^\vee(\mathbb{Q})_{{\mathrm{tor}}}}.
$

By substituting what we computed above, we obtain $ 3^2 \mid \char93 {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A)$ . Since $ L(A_{/\mathbb{Q}}, 1) \ne 0$ , [KL89] implies that $ {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A)$ is finite. By the nondegeneracy of the Cassels-Tate pairing, $ \char93  {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{...
...ontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(A^\vee/\mathbb{Q})$ . Thus, if the BSD conjectural formula is true then $ 3^2 \mid \char93  {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\f...
...ntfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A^\vee)$ . $ \qedsymbol$

We next observe that there are no visible elements of odd order for the embedding $ {A_{23}}_{/\mathbb{Q}} \hookrightarrow {J_0(767)}_{/\mathbb{Q}}$ .

Lemma 6.1.2   Any element of $ {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A_{23})$ which is visible in $ J_0(767)$ has order a power of $ 2$ .

Proof. Since $ m_{A_{23}} = 2^{34}$ , [AS05, Prop. 3.15] implies that any element of $ {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A_{23})$ that is visible in $ J_0(767)$ has order a power of $ 2$ . $ \qedsymbol$

Finally, we use Theorem 5.4.2 to prove the existence of non-trivial elements of order 3 in $ {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A_{23})$ which are invisible at level $ 767$ , but become visible at higher level. In particular, we prove unconditionally that $ 3 \mid \char93 {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A_{23})$ which provides evidence for the Birch and Swinnerton-Dyer conjectural formula.

Proposition 6.1.3   There is an element of order 3 in $ {\mbox{{\fontencoding{OT2}\fontfamily{wncyr}\fontseries{m}\fontshape{n}\selectfont Sh}}}(\mathbb{Q}, A_{23})$ which is not visible in $ J_0(767)$ but is strongly visible in $ J_0(2 \cdot 767)$ .

Proof. Let $ A=A_{23}$ , and note that $ A$ has rank 0 , since $ L(A_{/\mathbb{Q}},1)\neq 0$ . Using Cremona's database [Cre] we find that the elliptic curve

$\displaystyle E:\qquad y^2 + xy = x^3 - x^2 + 5x + 37
$

has conductor $ 2\cdot 767$ and Mordell-Weil group $ E(\mathbb{Q}) = \mathbb{Z}\oplus \mathbb{Z}$ . Also

$\displaystyle c_2 = 2, c_{13} = 2, c_{59} = 1,
\overline{c}_2 = 6, \overline{c}_{13} = 2, \overline{c}_{59}=1.
$

We apply Theorem 5.4.2 with $ \ell=3$ and $ p=2$ . Since $ E$ does not admit any rational $ 3$ -isogeny (by [Cre]), Hypothesis 3 is satisfied. The level is square free and the modular degree of $ A$ is a power of $ 2$ , so Hypothesis 2 is satisfied.

We have $ a_3(E) = -3$ . Using Magma we find

$\displaystyle \det(T_3\vert _{\overline{W}} - (-3)) \equiv 1\pmod{3},
$

which verifies the noncongruence hypothesis and completes the proof.

$ \qedsymbol$

William Stein 2006-06-21