Abstract
Let \(\mathcal {Q}_D\) be the set of moduli points on Siegel’s modular threefold whose corresponding principally polarized abelian surfaces have quaternionic multiplication by a maximal order \(\mathcal {O}\) in an indefinite quaternion algebra of discriminant D over \(\mathbb {Q}\) such that the Rosati involution coincides with a positive involution of the form \(\alpha \mapsto \mu ^{-1}\overline{\alpha }\mu \) on \(\mathcal {O}\) for some \(\mu \in \mathcal {O}\) with \(\mu ^2+D=0\). In this paper, we first give a formula for the number of irreducible components in \(\mathcal {Q}_D\), strengthening an earlier result of Rotger. Then for each irreducible component of genus 0, we determine its rational parameterization in terms of a Hauptmodul of the associated Shimura curve.
Similar content being viewed by others
References
Baba, S., Granath, H.: Genus 2 curves with quaternionic multiplication. Can. J. Math. 60(4), 734–757 (2008)
Bolza, O.: On binary sextics with linear transformations into themselves. Am. J. Math. 10(1), 47–70 (1887)
Bonfanti, M.A., van Geemen, B.: Abelian surfaces with an automorphism and quaternionic multiplication. Can. J. Math. 68(1), 24–43 (2016)
Borcherds, R.E.: Automorphic forms with singularities on Grassmannians. Invent. Math. 132(3), 491–562 (1998)
Borcherds, R.E.: Reflection groups of Lorentzian lattices. Duke Math. J. 104(2), 319–366 (2000)
Bruinier, J.H.: Borcherds products on O(2, \(l\)) and Chern classes of Heegner divisors. Lecture Notes in Mathematics, vol. 1780. Springer, Berlin (2002)
Cardona, G.: \({\mathbb{Q}}\)-curves and abelian varieties of \(\rm GL_2\)-type from dihedral genus 2 curves. In: Modular Curves and Abelian Varieties, vol. 224 of Progr. Math., pp. 45–52. Birkhäuser, Basel (2004)
Cardona, G., Quer, J.: Curves of genus 2 with group of automorphisms isomorphic to \(D_8\) or \(D_{12}\). Trans. Am. Math. Soc. 359(6), 2831–2849 (2007)
Cox, D.A.: Primes of the form \(x^2 + ny^2\). A Wiley-Interscience Publication, Wiley, New York (1989) (Fermat, class field theory and complex multiplication)
Elkies, N., Kumar, A.: K3 surfaces and equations for Hilbert modular surfaces. Algebra Number Theory 8(10), 2297–2411 (2014)
Errthum, E.: Singular moduli of Shimura curves. Can. J. Math. 63(4), 826–861 (2011)
González, J., Guàrdia, J.: Genus two curves with quaternionic multiplication and modular Jacobian. Math. Comput. 78(265), 575–589 (2009)
González, J., Molina, S.: The kernel of Ribet’s isogeny for genus three Shimura curves. J. Math. Soc. Jpn. 68(2), 609–635 (2016)
Goren, E.Z., Lauter, K.E.: Class invariants for quartic CM fields. Ann. Inst. Fourier (Grenoble) 57(2), 457–480 (2007)
Guo, J.-W., Yang, Y.: Equations of hyperelliptic Shimura curves. Compos. Math. 153(1), 1–40 (2017)
Hashimoto, K.: Explicit form of quaternion modular embeddings. Osaka J. Math. 32, 533–546 (1995)
Hashimoto, K., Murabayashi, N.: Shimura curves as intersections of Humbert surfaces and defining equations of QM-curves of genus two. Tohoku Math. J. 47(2), 271–296 (1995)
Igusa, J.: Arithmetic variety of moduli for genus two. Ann. Math. 2(72), 612–649 (1960)
Igusa, J.: On Siegel modular forms of genus two. Am. J. Math. 84, 175–200 (1962)
Igusa, J.: Modular forms and projective invariants. Am. J. Math. 89, 817–855 (1967)
Kudla, S.S.: Integrals of Borcherds forms. Compos. Math. 137(3), 293–349 (2003)
Lang, S.: Introduction to algebraic and Abelian functions, vol. 89 of Graduate Texts in Mathematics, 2nd edn. Springer, New York, Berlin (1982)
Mestre, J.F.: Construction de courbes de genre \(2\) à partir de leurs modules. In: Effective methods in Algebraic Geometry (Castiglioncello, 1990), vol. 94 of Progr. Math., pp. 313–334. Birkhäuser, Boston (1991)
Mumford, D.: Abelian varieties. In: Tata Institute of Fundamental Research Studies in Mathematics, vol. 5. Published for the Tata Institute of Fundamental Research, Bombay. Hindustan Book Agency, New Delhi (2008) [With appendices by C.P. Ramanujam and Yuri Manin. Corrected reprint of the second (1974) edition]
Nelson, P.D.: Evaluating modular forms on Shimura curves. Math. Comput. 84(295), 2471–2503 (2015)
Ogg, A.P.: Real points on Shimura curves. In: Arithmetic and Geometry, vol. I, volume 35 of Progr. Math., pp. 277–307. Birkhäuser, Boston (1983)
Pollak, B.: The equation \(\bar{tat}=b\) in a composition algebra. Duke Math. J. 29, 225–230 (1962)
Rotger, V.: Abelian varieties with quaternionic multiplication and their moduli. Thesis (Ph.D.)–Universitat de Barcelona (2002)
Rotger, V.: Quaternions, polarization and class numbers. J. Reine Angew. Math. 561, 177–197 (2003)
Rotger, V.: Modular Shimura varieties and forgetful maps. Trans. Am. Math. Soc. 356(4), 1535–1550 (2004)
Rotger, V.: Shimura curves embedded in Igusa’s threefold. In: Modular Curves and Abelian Varieties, vol. 224 of Progr. Math., pp. 263–276. Birkhäuser, Basel (2004)
Runge, B.: Endomorphism rings of abelian surfaces and projective models of their moduli spaces. Tohoku Math. J. 51(3), 283–303 (1999)
Schofer, J.: Borcherds forms and generalizations of singular moduli. J. Reine Angew. Math. 629, 1–36 (2009)
Shimizu, H.: Theta series and automorphic forms on \({\rm GL}_{2}\). J. Math. Soc. Jpn. 24, 638–683 (1972)
Shimura, G.: Construction of class fields and zeta functions of algebraic curves. Ann. Math. 2(85), 58–159 (1967)
Vignéras, M.F.: Arithmétique des algèbres de quaternions, vol. 800 of Lecture Notes in Mathematics. Springer, Berlin (1980)
Watson, T.C.: Rankin triple products and quantum chaos. ProQuest LLC, Ann Arbor, MI (2002) [Thesis (Ph.D.)–Princeton University]
Yang, Y.: Special values of hypergeometric functions and periods of CM elliptic curves. Trans. Am. Math. Soc. 370(9), 6433–6467 (2018)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Yi-Hsuan Lin and Yifan Yang would like to thank the referee for thorough reading the manuscript and providing valuable comments that improve the exposition of the paper substantially. Yi-Hsuan Lin and Yifan Yang were partially supported by Grant 106-2115-M-002-009-MY3 of the Ministry of Science and Technology, Taiwan (R.O.C.).
Appendices
Appendix A: List of Shimura curves and their quadratic forms
Here we list all Shimura curves of discriminant \(<100\) and Shimura curves of genus zero and discriminant \(>100\) on \(\mathscr {A}_2\), characterized by their quadratic forms. Here D is the discriminant of the quaternion algebra, k is the numbering of the curves, W is the stable group of the Shimura curve, as defined in Definition 11, and g is the genus of the Shimura curve. We let [a, b, c] represent the quadratic form \(ax^2+bxy+cy^2\). The quadratic forms are enumerated first in the ascending order of b and then in the ascending order of a.
Appendix B: Modular parameterizations of Shimura curves
Let \(s_k\), \(k=2,3,5,6\), be the Siegel modular forms defined in (14). In this appendix, we give modular parameterizations for Shimura curves of genus zero on \(\mathscr {A}_2\). Here we only list Shimura curves of discriminant D up to 50 and refer the reader to the arXiv version of the paper for a complete list.
Our choices of Hauptmoduls for \(X_0^D(1)/W_D\) and \(X_0^D(1)/w_D\) are given in 2 and 3, respectively. Note that the description of Hauptmoduls is given by specifying the values of the Hauptmodul at three certain CM-points \(z_{d_1}\), \(z_{d_2}\), and \(z_{d_3}\). In the case of \(X_0^D(1)/W_D\), this uniquely determines the Hauptmodul since for each \(d_i\), there is only one CM-point of discriminant \(d_i\). In the case of \(X_0^D(1)/w_D\), we have two different CM-points of discriminant \(d_3\). Thus, there are two possible choices of Hauptmoduls, related by the Atkin–Lehner involution \(w_q\), q|D. In the table, we also describe their relations (Tables 1, 2, 3).
\(\mathfrak {X}_{6}^1\)\(\quad [ 5, 2, 5 ]\) | |
\(s_2\) | j |
\(s_3\) | \(j^2\) |
\(s_5\) | \(j^3\) |
\(s_6\) | \(j^3(j+1)\) |
\(\mathfrak {X}_{10}^1\)\(\quad [ 5, 0, 8 ]\) | |
\(s_2\) | \(5^2(j-1)^2\) |
\(s_3\) | \(2^1(j-1)^2(49j+5)\) |
\(s_5\) | \(2^23^5(j-1)^4\) |
\(s_6\) | \(3^5(j-1)^4(3j^2+10j+35)\) |
\(\mathfrak {X}_{14}^1\)\(\quad [ 5, 4, 12 ]\) | |
\(s_2\) | \((3j+5)^2\) |
\(s_3\) | \(7^1(27j^2+36j+14)\) |
\(s_5\) | \(3^5(j-1)^2j^2\) |
\(s_6\) | \(3^5(3j^6-6j^5+4j^4+4j^3+j^2-6j+3)\) |
\(\mathfrak {X}_{15}^1\)\(\quad [ 5, 0, 12 ]\) | |
\(s_2\) | \(5^2\) |
\(s_3\) | \(4j^2+64j+121\) |
\(s_5\) | \(2^6(j-1)^2j\) |
\(s_6\) | \(2^4(j^4+72j^3+174j^2+8j+1)\) |
\(\mathfrak {X}_{15}^2\)\(\quad [ 8, 4, 8 ]\) | |
\(s_2\) | \(5^1(4j+1)\) |
\(s_3\) | \(4j^2+190j-5\) |
\(s_5\) | \(2^4(j-1)^2(j+3)\) |
\(s_6\) | \(2^4(j^4+15j^3+105j^2+125j+10)\) |
\(\mathfrak {X}_{21}^1\)\(\quad [ 5, 2, 17 ]\) | |
\(s_2\) | \((5j^2-16j-5)^2\) |
\(s_3\) | \(7^1(j-1)(j+1)(27j^4-144j^3+394j^2-624j+603)\) |
\(s_5\) | \(2^{12}(j-1)(j+1)(j+3)^2(j^2-5)^2\) |
\(s_6\) | \(2^{12}(j^{12}+6j^{11}-9j^{10}-108j^9-28j^8+708j^7+406j^6-2016j^5+17j^4+3846j^3-2637j^2 -2436j+6346)\) |
\(\mathfrak {X}_{22}^1\)\(\quad [ 8, 8, 13 ]\) | |
\(s_2\) | \((j-1)^2(9j+64)\) |
\(s_3\) | \((j-1)^2(351j^2-782j+539)\) |
\(s_5\) | \(3^5(j-1)^4j(j+1)^2\) |
\(s_6\) | \(3^5(j-1)^4(3j^5+5j^4+23j^3+10j^2+4j+3)\) |
\(\mathfrak {X}_{26}^1\)\(\quad [ 8, 0, 13 ]\) | |
\(s_2\) | \(25j^2-306j+1305\) |
\(s_3\) | \(2^1(49j^3+27j^2-5913j+23085)\) |
\(s_5\) | \(2^23^5(j-1)^2(j+3)^2\) |
\(s_6\) | \(3^5(3j^6-2j^5+13j^4-92j^3+2733j^2+7038j+2595)\) |
\(\mathfrak {X}_{26}^2\)\(\quad [ 5, 2, 21 ]\) | |
\(s_2\) | \((5j^2+12j+15)^2\) |
\(s_3\) | \(2^1(7j^2+12j+9)(7j^4+60j^3+144j^2+108j+297)\) |
\(s_5\) | \(2^23^5(j-3)^2(j-1)^2(j^2-5)^2\) |
\(s_6\) | \(3^5(3j^{12}-24j^{11}+22j^{10}+280j^9-651j^8-880j^7+3348j^6-848j^5-2819j^4+3336j^3+2166j^2-34632j+42987)\) |
\(\mathfrak {X}_{33}^1\)\(\quad [ 8, 4, 17 ]\) | |
\(s_2\) | \(288j^6-360j^5-295j^4+240j^3+238j^2+120j+25\) |
\(s_3\) | \((j-1)(j+1)(324j^8-4536j^7+9216j^6+4284j^5+3373j^4+4704j^3+3350j^2+924j+121)\) |
\(s_5\) | \(2^6(j-1)j^2(j+1)(3j-1)^2(3j^2+1)^2(3j^3+7j^2-3j+1)^2\) |
\(s_6\) | \(2^4(6561j^{20}+26244j^{19}+275562j^{18}+26244j^{17}+279693j^{16}+863136j^{15}-190440j^{14}-779904j^{13}+700018j^{12}-123384j^{11}-72484j^{10}+20328j^9+35570j^8-28416j^7+12888j^6-3936j^5+1133j^4-300j^3+74j^2-12j+1)\) |
\(\mathfrak {X}_{34}^1\)\(\quad [ 5, 4, 28 ]\) | |
\(s_2\) | \(j^2(11j^2-20j+1)^2\) |
\(s_3\) | \(j^2(27j^8-926j^7+9096j^6-10998j^5+8186j^4-4122j^3+1032j^2-82j+27)\) |
\(s_5\) | \(3^5j^4(j+1)^2(j+3)^2(j^2+2j-1)^2(j^2+4j-1)^2\) |
\(s_6\) | \(3^5j^4(3j^{16}+68j^{15}+683j^{14}+3836j^{13}+12666j^{12}+23708j^{11}+21565j^{10}+6724j^9+8838j^8+13276j^7-12811j^6-9628j^5+13338j^4-5180j^3+771j^2-36j+3)\) |
\(\mathfrak {X}_{35}^1\)\(\quad [ 5, 0, 28 ]\) | |
\(s_2\) | \(5^2(3j^2+10j+51)^2\) |
\(s_3\) | \(7^1(27j^2+178j+243)(27j^4-36j^3-502j^2-1044j+9747)\) |
\(s_5\) | \(2^{14}3^5(j-1)^2j(j^3+13j^2+51j-1)^2\) |
\(s_6\) | \(2^{12}3^5(3j^{12}+96j^{11}+2014j^{10}+35808j^9+389869j^8+2509312j^7+9837156j^6+21020288j^5+17081069j^4-730208j^3+186334j^2-96j+3)\) |
\(\mathfrak {X}_{35}^2\)\(\quad [ 12, 4, 12]\) | |
\(s_2\) | \(5^1(45j^4+876j^3-2002j^2-3924j+909)\) |
\(s_3\) | \(7^1(729j^6-9126j^5+34119j^4+490604j^3+124119j^2+642330j+27945)\) |
\(s_5\) | \(2^{12}3^5(j+3)^2(j+7)(j^3+13j^2+51j-1)^2\) |
\(s_6\) | \(2^{12}3^5(3j^{12}+141j^{11}+3019j^{10}+39267j^9+339240j^8+1967522j^7+7421414j^6+16970566j^5+20814403j^4+14639425j^3+19460895j^2+2222055j+8130)\) |
\(\mathfrak {X}_{35}^3\)\(\quad [ 12, 8, 13]\) | |
\(s_2\) | \(5^1(45j^8+288j^7+1356j^6+1824j^5+5390j^4+1632j^3+13644j^2-3744j+45)\) |
\(s_3\) | \(7^1(729j^{12}+3888j^{11}+18090j^{10}+79056j^9+277623j^8+649440j^7+1230476j^6+819360j^5+1037367j^4-1563408j^3+1089450j^2+11664j+16281)\) |
\(s_5\) | \(2^{12}3^5(j-1)^2(j+1)^2(j^2-2j+5)^2(j^2+3)^2(j^3-j^2+7j+1)^2\) |
\(s_6\) | \(2^{12}3^5(3j^{24}-18j^{23}+147j^{22}-576j^{21}+2635j^{20}-7266j^{19}+23897j^{18}-48084j^{17}+132420j^{16}-194660j^{15}+536174j^{14}-563952j^{13}+1917622j^{12}-1570836j^{11}+6456226j^{10}-4274680j^9+15283435j^8-4052682j^7+16219183j^6+6354768j^5+7786095j^4+4437942j^3+1954773j^2-79956j+19038)\) |
\(\mathfrak {X}_{38}^1\)\(\quad [ 12, 4, 13]\) | |
\(s_2\) | \(9j^6-60j^5+346j^4-240j^3+153j^2-180j+36\) |
\(s_3\) | \(675j^8-2016j^7+2600j^6-8064j^5+11610j^4-6048j^3+3240j^2+243\) |
\(s_5\) | \(3^5(j+1)^2(j^2+1)^2(j^2+3)^2(j^2+j+2)^2\) |
\(s_6\) | \(3^5(3j^{18}+12j^{17}+61j^{16}+160j^{15}+525j^{14}+1068j^{13}+2612j^{12}+4032j^{11}+7533j^{10}+8548j^9+12532j^8+9984j^7+11971j^6+6180j^5+6792j^4+2208j^3+2592j^2+576j+435)\) |
\(\mathfrak {X}_{38}^2\)\(\quad [ 8, 8, 21 ]\) | |
\(s_2\) | \(9j^3+46j^2+393j+576\) |
\(s_3\) | \(81j^4-836j^3-6822j^2-13068j-13851\) |
\(s_5\) | \(-3^5(j-1)^2j(j+1)^2(3j+1)^2\) |
\(s_6\) | \(3^5(3j^9+153j^8+1077j^7+2848j^6+4149j^5+2872j^4+987j^3+172j^2+24j+3)\) |
\(\mathfrak {X}_{39}^1\)\(\quad [ 5, 4, 32 ]\) | |
\(s_2\) | \((5j^4+4j^3-10j^2+4j+5)^2\) |
\(s_3\) | \(189j^{12}-180j^{11}+514j^{10}-708j^9-45j^8+24j^7-68j^6+2328j^5-45j^4-2148j^3-158j^2+684j+189\) |
\(s_5\) | \(2^{10}(j-2)^2j^2(j^2-2j-1)^2(j^2-j-1)^2(j^3-j^2-2j-1)^2\) |
\(s_6\) | \(2^{10}(4j^{24}-60j^{23}+363j^{22}-1056j^{21}+1088j^{20}+1680j^{19}-4884j^{18}+210j^{17}+8824j^{16}-3396j^{15}-10122j^{14}+3522j^{13}+8843j^{12}-348j^{11}-5640j^{10}-2082j^9+2134j^8+1632j^7\) |
\(-537j^6-642j^5+275j^4+492j^3+228j^2+48j+4)\) | |
\(\mathfrak {X}_{39}^2\)\(\quad [ 8, 4, 20 ]\) | |
\(s_2\) | \(25j^8-40j^7-4j^6-200j^5-58j^4+520j^3+556j^2+200j+25\) |
\(s_3\) | \(189j^{12}-684j^{11}-998j^{10}+5340j^9+2139j^8-13752j^7-9308j^6+9384j^5+17931j^4+14148j^3+6562j^2+1692j+189\) |
\(s_5\) | \(2^{10}(j-1)^2j^2(j^2-j-1)^2(j^3-2j^2+j+1)^2(j^4-j^3-j^2+j+1)\) |
\(s_6\) | \(2^{10}(4j^{24}-48j^{23}+243j^{22}-639j^{21}+773j^{20}+219j^{19}-1875j^{18}+1560j^{17}+1642j^{16}-3300j^{15}-255j^{14}+3855j^{13}-1018j^{12}-3420j^{11}+1593j^{10}+2352j^9-1301j^8-1260j^7+639j^6+516j^5-172j^4-147j^3+15j^2+24j+4)\) |
Appendix C: Mestre obstructions
In [23], Mestre gave an algorithm to generate a hyperelliptic curve of genus 2 with given Igusa invariants \(J=[J_2,J_4,J_6,J_{10}]\). In the process, he considered a certain ternary quadratic form
constructed from J. He showed that if L(J) is degenerate, then there is always a curve C of genus 2 defined over \(\mathbb {Q}(J):=\mathbb {Q}(J_2,J_4,J_6,J_{10})\). In such a case, the curve C has a nontrivial automorphism different from the hyperelliptic involution. (For the case of curves over \(\mathbb {C}\), this means that the Jacobian of C lies on the Humbert surface of discriminant 4. See [2].) Furthermore, he showed that when L(J) is nondegenerate, there is a curve C of genus 2 defined over a field K with Igusa invariants J if and only if L(J) is isotropic over K. In other words, there is a quaternion algebra \(\mathcal {B}\) over \(\mathbb {Q}(J)\) such that there is a curve over K with Igusa invariants J if and only if K splits \(\mathcal {B}\). (In the case when L(J) is diagonal, say, \(L(J)=a_1x_1^2+a_2x_2^2+a_3x_3^2\), \(\mathcal {B}\) is simply \(\left( \frac{-a_1a_3,-a_2a_3}{\mathbb {Q}(J)}\right) \)). In literature, this quaternion algebra \(\mathcal {B}\) is called the Mestre obstruction for J. In the case of the unique Shimura curve \(\mathfrak {X}_6\) of discriminant 6, using the parameterization given in (2), Baba and Granath [1] exhibited a matrix \(M\in M(3,\mathbb {Z}[j])\) such that
Therefore, for a point on \(\mathfrak {X}_6\) that is not on \(H_4\), the Mestre obstruction is \(\left( \frac{-6j,-2(27j+16)}{\mathbb {Q}(j)}\right) \). For the case of \(D=10\), Baba and Granath [1] found that the Mestre obstruction is \(\left( \frac{-10j,-5(2j+25)}{\mathbb {Q}(j)}\right) \). (Note that their choice of Hauptmodul is different from ours.) In this section, we conduct a similar computation and determine Mestre obstructions for Shimura curves of genus 0 and discriminant less than 50 (80 in the arXiv version). The results are given in Table 4.
Remark 44
Note that when \(X_0^D(1)/w_D\) has genus 0 and j is a Hauptmodul for \(X_0^D(1)/w_D\), a canonical model for \(X_0^D(1)\) has the form \(y^2=f(x)\), where the roots of f(x) are the values of j at the fixed points of the Atkin-Lehner involution \(w_D\). Our computation shows that in all cases where the Shimura curve \(\mathfrak {X}\) is isomorphic to \(X_0^D(1)/w_D\), the Mestre obstruction for \(\mathfrak {X}\) is given by
where m is an integer such that \(r^2m\) is representable by the quadratic form associated to \(\mathfrak {X}\) for some rational number r. For instance, the quadratic form for \(\mathfrak {X}_{51}^2\) is \(5x^2+2xy+41y^2\), which clearly represents 5. Also, a canoncal model for \(X_0^{51}(1)\) is \(y^2=f(x)\), where \(f(x)=-(x^2+3)(243x^6+235x^4-31x^2+1)\) (see [13, 15]). Our computation shows that the Mestre obstruction for \(\mathfrak {X}_{51}^2\) is \(\left( \frac{-51,5f(j)}{\mathbb {Q}(j)}\right) \). If this phenomenon holds in general, there should be a deep arithmetic meaning.
Note also that it is well-known that the Shimura curve \(X_0^D(1)\) has no real points when \(D>1\) (see [26]). In other words, the polynomial f(x) above is negative for any real x. It follows that the quaternion algebra \(\left( \frac{-D,mp(j)}{\mathbb {Q}}\right) \) always ramifies at the infinite place whenever \(j\in \mathbb {R}\). Therefore, we have the following proposition.
Proposition 45
Let \(\mathfrak {X}\) be a Shimura curve in Table 4 that is isomorphic to \(X_0^D(1)/w_D\). (I.e., the quadratic form associated to \(\mathfrak {X}\) is not ambiguous.) Then there are only a finite number of isomorphism classes of genus 2 curves over \(\mathbb {R}\) such that their Jacobians lie on \(\mathfrak {X}\). To be more precise, these exceptional moduli points are the real points that lie on the intersection of \(\mathfrak {X}\) and \(H_4\), but not on \(H_1\).
Example 46
Consider \(\mathfrak {X}=\mathfrak {X}_{14}^1\). Using (15) and the modular parameterization in Appendix B, we find that the only real points lying on \(\mathfrak {X}\cap H_4\) have j-values \(j=\infty \),0,\(\pm 1\),5 / 9,\(\pm \sqrt{5}\). Among these points, the points corresponding to \(j=\infty ,0,1\) also lie on \(H_1\). Thus, there are precisely four genus 2 curves over \(\mathbb {R}\) whose Jacobians belong to \(\mathfrak {X}\), two of them defined over \(\mathbb {Q}\) and the other two defined over \(\mathbb {Q}(\sqrt{5})\). The point with \(j=-1\) is a CM-point of discriminant \(-11\) with
or equivalently,
Its ternary quadratic form
represents \(4x^2+4xy+4y^2\). Thus, it also lies on the modular curve \(\mathfrak {Y}_3'\), defined in Sect. 5. Since \(\mathfrak {Y}_3'\) parameterizes curves of genus 2 with an automorphism group containing the dihedral group \(D_6\) of order 12 (see [7, Theorem 2.1]), a curve of genus 2 with these invariants can be obtained using a result of Cardona and Quer [8, Proposition 2.2]. We find that
is a curve with \([s_2,s_3,s_5,s_6]=[4,35,972,4617]\).
The point with \(j=5/9\) is a CM-point of discriminant \(-43\) with
or equivalently,
A curve of genus 2 over \(\mathbb {Q}\) with these invariants is
where
An extra involution is given by \((x,y)\mapsto (-43/x,\sqrt{-43^3}y/x^3)\).
Rights and permissions
About this article
Cite this article
Lin, YH., Yang, Y. Quaternionic loci in Siegel’s modular threefold. Math. Z. 295, 775–819 (2020). https://doi.org/10.1007/s00209-019-02372-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00209-019-02372-z