|
|
(2 intermediate revisions by one other user not shown) |
Line 1: |
Line 1: |
− | An [[Abelian variety|Abelian variety]] of dimension two, i.e. a complete connected group variety of dimension two over a field <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100401.png" />. The group law is commutative. In the sequel, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100402.png" /> is assumed to be algebraically closed (cf. [[Algebraically closed field|Algebraically closed field]]).
| + | <!-- |
| + | a1100401.png |
| + | $#A+1 = 261 n = 12 |
| + | $#C+1 = 261 : ~/encyclopedia/old_files/data/A110/A.1100040 Abelian surface |
| + | Automatically converted into TeX, above some diagnostics. |
| + | Please remove this comment and the {{TEX|auto}} line below, |
| + | if TeX found to be correct. |
| + | --> |
| | | |
− | In the classification of algebraic surfaces, Abelian surfaces are exactly the smooth complete surfaces <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100403.png" /> with [[Kodaira dimension|Kodaira dimension]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100404.png" />, [[Geometric genus|geometric genus]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100405.png" /> and [[Irregularity|irregularity]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100406.png" />.
| + | {{TEX|auto}} |
| + | {{TEX|done}} |
| | | |
− | For an Abelian surface <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100407.png" />, the dual Abelian variety <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100408.png" /> is again an Abelian surface. An invertible [[Sheaf|sheaf]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a1100409.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004010.png" /> defines the homomorphism <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004011.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004012.png" />. The homomorphism <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004013.png" /> depends only on the algebraic equivalence class of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004014.png" />. The invertible sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004015.png" /> is ample (cf. [[Ample sheaf|Ample sheaf]]) if and only if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004016.png" /> is an [[Isogeny|isogeny]] (i.e., <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004017.png" /> is surjective and has finite kernel) and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004018.png" />. In this case, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004019.png" /> with a positive integer <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004020.png" /> and the [[Riemann–Roch theorem|Riemann–Roch theorem]] says that
| + | An [[Abelian variety|Abelian variety]] of dimension two, i.e. a complete connected group variety of dimension two over a field $ k $. |
| + | The group law is commutative. In the sequel, $ k $ |
| + | is assumed to be algebraically closed (cf. [[Algebraically closed field|Algebraically closed field]]). |
| | | |
− | <table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004021.png" /></td> </tr></table>
| + | In the classification of algebraic surfaces, Abelian surfaces are exactly the smooth complete surfaces $ A $ |
| + | with [[Kodaira dimension|Kodaira dimension]] $ \kappa = 0 $, |
| + | [[Geometric genus|geometric genus]] $ p _ {g} = h ^ {2} ( A, {\mathcal O} _ {A} ) =1 $ |
| + | and [[Irregularity|irregularity]] $ q = h ^ {1} ( A, {\mathcal O} _ {A} ) = 2 $. |
| | | |
− | where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004022.png" /> denotes the self-intersection number. Every Abelian surface admits an ample invertible sheaf and hence is projective (cf. [[Projective scheme|Projective scheme]]).
| + | For an Abelian surface $ A $, |
| + | the dual Abelian variety $ {\widehat{A} } = { \mathop{\rm Pic} } ^ {0} ( A ) $ |
| + | is again an Abelian surface. An invertible [[Sheaf|sheaf]] $ L $ |
| + | on $ A $ |
| + | defines the homomorphism $ {\phi _ {L} } : A \rightarrow { {\widehat{A} } } $, |
| + | $ a \mapsto t _ {a} ^ {*} L \otimes L ^ {- 1 } $. |
| + | The homomorphism $ \phi _ {L} $ |
| + | depends only on the algebraic equivalence class of $ L $. |
| + | The invertible sheaf $ L $ |
| + | is ample (cf. [[Ample sheaf|Ample sheaf]]) if and only if $ \phi _ {L} $ |
| + | is an [[Isogeny|isogeny]] (i.e., $ \phi _ {L} $ |
| + | is surjective and has finite kernel) and $ h ^ {0} ( A,L ) \neq0 $. |
| + | In this case, $ { \mathop{\rm deg} } \phi _ {L} = d ^ {2} $ |
| + | with a positive integer $ d $ |
| + | and the [[Riemann–Roch theorem|Riemann–Roch theorem]] says that |
| | | |
− | A polarization <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004023.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004024.png" /> is the algebraic equivalence class <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004025.png" /> of an ample invertible sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004026.png" />. The degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004027.png" /> of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004028.png" /> is by definition <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004029.png" />. An Abelian surface <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004030.png" /> together with a polarization <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004031.png" /> is a polarized Abelian surface. A principal polarization is a polarization of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004032.png" />. A principally polarized Abelian surface <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004033.png" /> is either the [[Jacobi variety|Jacobi variety]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004034.png" /> of a smooth projective curve <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004035.png" /> of genus <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004036.png" />, and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004037.png" /> is the class of the theta divisor, or <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004038.png" /> is the product of two elliptic curves (Abelian varieties of dimension one, cf. also [[Elliptic curve|Elliptic curve]]) with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004039.png" /> the product polarization.
| + | $$ |
| + | h ^ {0} ( A,L ) = { |
| + | \frac{1}{2} |
| + | } ( L ^ {2} ) = d, |
| + | $$ |
| | | |
− | If the degree of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004040.png" /> is prime to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004041.png" />, then <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004042.png" /> is said to be a separable polarization and the kernel of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004043.png" /> is isomorphic to the group <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004044.png" />, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004045.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004046.png" /> are positive integers such that <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004047.png" /> divides <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004048.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004049.png" />. The pair <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004050.png" /> is the type of the polarized Abelian surface <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004051.png" />.
| + | where $ ( L ^ {2} ) $ |
| + | denotes the self-intersection number. Every Abelian surface admits an ample invertible sheaf and hence is projective (cf. [[Projective scheme|Projective scheme]]). |
| | | |
− | A polarization <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004052.png" /> of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004053.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004054.png" /> defines a polarization <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004055.png" /> on the dual Abelian surface <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004056.png" />. The polarization <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004057.png" /> is again of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004058.png" /> and it is characterized by each of the following two equivalent properties: | + | A polarization $ \lambda $ |
| + | on $ A $ |
| + | is the algebraic equivalence class $ [ L ] $ |
| + | of an ample invertible sheaf $ L $. |
| + | The degree $ { \mathop{\rm deg} } \lambda $ |
| + | of $ \lambda $ |
| + | is by definition $ d = \sqrt { { \mathop{\rm deg} } \phi _ {L} } $. |
| + | An Abelian surface $ A $ |
| + | together with a polarization $ \lambda $ |
| + | is a polarized Abelian surface. A principal polarization is a polarization of degree $ 1 $. |
| + | A principally polarized Abelian surface $ ( A, \lambda ) $ |
| + | is either the [[Jacobi variety|Jacobi variety]] $ J ( H ) $ |
| + | of a smooth projective curve $ H $ |
| + | of genus $ 2 $, |
| + | and $ \lambda = \theta $ |
| + | is the class of the theta divisor, or $ A $ |
| + | is the product of two elliptic curves (Abelian varieties of dimension one, cf. also [[Elliptic curve|Elliptic curve]]) with $ \lambda $ |
| + | the product polarization. |
| | | |
− | <table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004059.png" /></td> </tr></table>
| + | If the degree of $ \lambda = [ L ] $ |
| + | is prime to $ { \mathop{\rm char} } ( k ) $, |
| + | then $ \lambda $ |
| + | is said to be a separable polarization and the kernel of $ \phi _ {L} $ |
| + | is isomorphic to the group $ ( \mathbf Z/d _ {1} \mathbf Z ) ^ {2} \times ( \mathbf Z/d _ {2} \mathbf Z ) ^ {2} $, |
| + | where $ d _ {1} $ |
| + | and $ d _ {2} $ |
| + | are positive integers such that $ d _ {1} $ |
| + | divides $ d _ {2} $ |
| + | and $ d _ {1} d _ {2} = { \mathop{\rm deg} } \lambda $. |
| + | The pair $ ( d _ {1} ,d _ {2} ) $ |
| + | is the type of the polarized Abelian surface $ ( A, \lambda ) $. |
| | | |
− | For a polarized Abelian surface <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004060.png" /> of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004061.png" />, the assignment <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004062.png" /> defines a rational mapping from <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004063.png" /> into the projective space of hyperplanes in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004064.png" />:
| + | A polarization $ \lambda = [ L ] $ |
| + | of type $ ( d _ {1} ,d _ {2} ) $ |
| + | on $ A $ |
| + | defines a polarization $ {\widehat \lambda } = [ {\widehat{L} } ] $ |
| + | on the dual Abelian surface $ {\widehat{A} } $. |
| + | The polarization $ {\widehat \lambda } $ |
| + | is again of type $ ( d _ {1} ,d _ {2} ) $ |
| + | and it is characterized by each of the following two equivalent properties: |
| | | |
− | <table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004065.png" /></td> </tr></table>
| + | $$ |
| + | \phi _ {L} ^ {*} {\widehat \lambda } = d _ {1} d _ {2} \lambda \iff \phi _ { {\widehat{L} } } \phi _ {L} = d _ {1} d _ {2} { \mathop{\rm id} } _ {A} . |
| + | $$ |
| | | |
− | If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004066.png" />, then <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004067.png" /> is everywhere defined. The Lefschetz theorem says that for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004068.png" /> the morphism <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004069.png" /> is an embedding. Suppose <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004070.png" />; then <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004071.png" /> with a polarization <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004072.png" /> of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004073.png" />. If the linear system <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004074.png" /> has no fixed components, then <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004075.png" /> is an embedding.
| + | For a polarized Abelian surface $ ( A, \lambda = [ L ] ) $ |
| + | of type $ ( d _ {1} ,d _ {2} ) $, |
| + | the assignment $ A \ni a \mapsto \{ {\sigma \in H ^ {0} ( A,L ) } : {\sigma ( a ) = 0 } \} \subset H ^ {0} ( A,L ) $ |
| + | defines a rational mapping from $ A $ |
| + | into the projective space of hyperplanes in $ H ^ {0} ( A,L ) $: |
| + | |
| + | $$ |
| + | {\varphi _ {L} } : A \rightarrow {\mathbf P ( H ^ {0} ( A,L ) ^ {*} ) } \simeq \mathbf P _ {k} ^ {d _ {1} d _ {2} - 1 } . |
| + | $$ |
| + | |
| + | If $ d _ {1} \geq 2 $, |
| + | then $ \varphi _ {L} $ |
| + | is everywhere defined. The Lefschetz theorem says that for $ d _ {1} \geq 3 $ |
| + | the morphism $ \varphi _ {L} $ |
| + | is an embedding. Suppose $ d _ {1} = 2 $; |
| + | then $ \lambda = 2 \mu $ |
| + | with a polarization $ \mu = [ M ] $ |
| + | of type $ ( 1, { {d _ {2} } / 2 } ) $. |
| + | If the linear system $ | M | $ |
| + | has no fixed components, then $ \varphi _ {L} $ |
| + | is an embedding. |
| | | |
| ==Complex Abelian surfaces.== | | ==Complex Abelian surfaces.== |
− | An Abelian surface over the field <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004076.png" /> of complex numbers is a [[Complex torus|complex torus]] | + | An Abelian surface over the field $ \mathbf C $ |
| + | of complex numbers is a [[Complex torus|complex torus]] |
| | | |
− | <table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004077.png" /></td> </tr></table>
| + | $$ |
| + | A = { {\mathbf C ^ {2} } / \Lambda } |
| + | $$ |
| | | |
− | (with a lattice <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004078.png" /> in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004079.png" />) admitting a polarization. A polarization <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004080.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004081.png" /> can be considered as a non-degenerate alternating form <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004082.png" />, the elementary divisors of which are given by the type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004083.png" /> of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004084.png" />. | + | (with a lattice $ \Lambda \simeq \mathbf Z ^ {4} $ |
| + | in $ \mathbf C ^ {2} $) |
| + | admitting a polarization. A polarization $ \lambda $ |
| + | on $ A $ |
| + | can be considered as a non-degenerate alternating form $ \Lambda \times \Lambda \rightarrow \mathbf Z $, |
| + | the elementary divisors of which are given by the type $ ( d _ {1} ,d _ {2} ) $ |
| + | of $ \lambda $. |
| | | |
− | In the sequel, the field <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004085.png" /> is assumed to be <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004086.png" />, although some of the following results are also valid for arbitrary algebraically closed fields. | + | In the sequel, the field $ k $ |
| + | is assumed to be $ \mathbf C $, |
| + | although some of the following results are also valid for arbitrary algebraically closed fields. |
| | | |
− | Suppose <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004087.png" /> is of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004088.png" /> and the linear system <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004089.png" /> has no fixed components. The Reider theorem states that for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004090.png" /> the invertible sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004091.png" /> is very ample if and only if there is no elliptic curve <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004092.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004093.png" /> with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004094.png" /> (see [[#References|[a14]]] and [[#References|[a10]]]). For arbitrary <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004095.png" /> there exist finitely many isogenies <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004096.png" /> of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004097.png" /> onto principally polarized Abelian surfaces (cf. also [[Isogeny|Isogeny]]). Suppose <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004098.png" /> with a symmetric invertible sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a11004099.png" /> (i.e., <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040100.png" />) and let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040101.png" /> be the unique divisor in the linear system <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040102.png" />. The divisor <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040103.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040104.png" /> defines a symmetric invertible sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040105.png" /> with class <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040106.png" /> and the covering <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040107.png" /> is étale of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040108.png" />. One distinguishes two cases: i) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040109.png" /> is smooth of genus <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040110.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040111.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040112.png" /> is smooth of genus <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040113.png" />; and ii) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040114.png" /> is the sum <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040115.png" /> of two elliptic curves with intersection number <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040116.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040117.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040118.png" /> is the sum <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040119.png" /> of two elliptic curves with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040120.png" />. | + | Suppose $ ( A, \lambda = [ L ] ) $ |
| + | is of type $ ( 1,d ) $ |
| + | and the linear system $ | L | $ |
| + | has no fixed components. The Reider theorem states that for $ d \geq 5 $ |
| + | the invertible sheaf $ L $ |
| + | is very ample if and only if there is no elliptic curve $ E $ |
| + | on $ A $ |
| + | with $ ( E \cdot L ) = 2 $( |
| + | see [[#References|[a14]]] and [[#References|[a10]]]). For arbitrary $ d \geq 1 $ |
| + | there exist finitely many isogenies $ f : {( A, \lambda ) } \rightarrow {( B, \theta ) } $ |
| + | of degree $ d $ |
| + | onto principally polarized Abelian surfaces (cf. also [[Isogeny|Isogeny]]). Suppose $ \theta = [ \Theta ] $ |
| + | with a symmetric invertible sheaf $ \Theta $( |
| + | i.e., $ ( -1 ) _ {A} ^ {*} \Theta \simeq \Theta $) |
| + | and let $ H $ |
| + | be the unique divisor in the linear system $ | \Theta | $. |
| + | The divisor $ C = f ^ {- 1 } ( H ) $ |
| + | on $ A $ |
| + | defines a symmetric invertible sheaf $ L = {\mathcal O} _ {A} ( C ) $ |
| + | with class $ [ L ] = \lambda $ |
| + | and the covering $ {f \mid _ {C} } : C \rightarrow H $ |
| + | is étale of degree $ d $. |
| + | One distinguishes two cases: i) $ H $ |
| + | is smooth of genus $ 2 $, |
| + | $ B = J ( H ) $ |
| + | and $ C $ |
| + | is smooth of genus $ d + 1 $; |
| + | and ii) $ H $ |
| + | is the sum $ E _ {1} + E _ {2} $ |
| + | of two elliptic curves with intersection number $ ( E _ {1} \cdot E _ {2} ) = 1 $, |
| + | $ B = E _ {1} \times E _ {2} $ |
| + | and $ C $ |
| + | is the sum $ F _ {1} + F _ {2} $ |
| + | of two elliptic curves with $ ( F _ {1} \cdot F _ {2} ) = d $. |
| | | |
− | In the following list, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040121.png" /> is a polarized Abelian surface of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040122.png" /> such that <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040123.png" /> admits no fixed components | + | In the following list, $ ( A, \lambda = [ L ] ) $ |
| + | is a polarized Abelian surface of type $ ( d _ {1} ,d _ {2} ) $ |
| + | such that $ | L | $ |
| + | admits no fixed components |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040124.png" />— The linear system <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040125.png" /> has exactly <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040126.png" /> base points. The blow-up <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040127.png" /> of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040128.png" /> in these points admits a morphism <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040129.png" /> induced by <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040130.png" />. The general fibre of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040131.png" /> is a smooth curve of genus <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040132.png" />. The curve <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040133.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040134.png" /> defining <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040135.png" /> as above is double elliptic: <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040136.png" /> with an elliptic curve <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040137.png" />, and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040138.png" /> is isomorphic to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040139.png" /> (see [[#References|[a3]]]). | + | Type $ ( 1,2 ) $— |
| + | The linear system $ | L | $ |
| + | has exactly $ 4 $ |
| + | base points. The blow-up $ {\widetilde{A} } $ |
| + | of $ A $ |
| + | in these points admits a morphism $ { {\widetilde \varphi } _ {L} } : { {\widetilde{A} } } \rightarrow {\mathbf P ^ {1} } $ |
| + | induced by $ \varphi _ {L} $. |
| + | The general fibre of $ {\widetilde \varphi } _ {L} $ |
| + | is a smooth curve of genus $ 3 $. |
| + | The curve $ C $ |
| + | on $ A $ |
| + | defining $ L $ |
| + | as above is double elliptic: $ C { \mathop \rightarrow \limits ^ { {2:1 }} } E $ |
| + | with an elliptic curve $ E $, |
| + | and $ A $ |
| + | is isomorphic to $ { {J ( C ) } / E } $( |
| + | see [[#References|[a3]]]). |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040140.png" />— <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040141.png" /> defines a <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040142.png" />-fold covering <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040143.png" /> ramified along a curve <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040144.png" /> of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040145.png" />. The general divisor in the linear system <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040146.png" /> is a smooth curve of genus <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040147.png" />. There are <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040148.png" /> isogenies <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040149.png" /> of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040150.png" /> onto principally polarized Abelian surfaces. In case i) the smooth genus-<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040151.png" /> curve <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040152.png" /> is double elliptic: <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040153.png" />, and the embedding of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040154.png" /> into the Jacobian <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040155.png" /> induces an exact sequence | + | Type $ ( 1,3 ) $— |
| + | $ L $ |
| + | defines a $ 6 $- |
| + | fold covering $ {\varphi _ {L} } : A \rightarrow {\mathbf P ^ {2} } $ |
| + | ramified along a curve $ R \subset \mathbf P ^ {2} $ |
| + | of degree $ 18 $. |
| + | The general divisor in the linear system $ | L | $ |
| + | is a smooth curve of genus $ 4 $. |
| + | There are $ 4 $ |
| + | isogenies $ f : {( A, \lambda ) } \rightarrow {( B, \theta ) } $ |
| + | of degree $ 3 $ |
| + | onto principally polarized Abelian surfaces. In case i) the smooth genus- $ 4 $ |
| + | curve $ C \in | L | $ |
| + | is double elliptic: $ C { \mathop \rightarrow \limits ^ { {2:1 }} } E $, |
| + | and the embedding of $ E $ |
| + | into the Jacobian $ J ( C ) $ |
| + | induces an exact sequence |
| | | |
− | <table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040156.png" /></td> </tr></table>
| + | $$ |
| + | 0 \rightarrow E \times E \rightarrow J ( C ) \rightarrow A \rightarrow 0. |
| + | $$ |
| | | |
− | The étale <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040157.png" />-fold covering <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040158.png" /> induces a morphism <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040159.png" /> with image <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040160.png" />, the dual Abelian surface of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040161.png" /> (see [[#References|[a7]]]). | + | The étale $ 3 $- |
| + | fold covering $ {f \mid _ {C} } : C \rightarrow H $ |
| + | induces a morphism $ {f ^ {*} } : {J ( H ) } \rightarrow {J ( C ) } $ |
| + | with image $ {\widehat{A} } $, |
| + | the dual Abelian surface of $ A $( |
| + | see [[#References|[a7]]]). |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040162.png" />— There are <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040163.png" /> isogenies <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040164.png" /> of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040165.png" /> onto principally polarized Abelian surfaces. If the curves <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040166.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040167.png" /> do not admit elliptic involutions compatible with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040168.png" />, then <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040169.png" /> is a birational morphism onto a singular octic <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040170.png" />. In the exceptional case, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040171.png" /> is a double covering of a singular quartic <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040172.png" />, which is birational to an elliptic scroll. In the first case the octic <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040173.png" /> is smooth outside the four coordinate planes of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040174.png" /> and touches the coordinate planes in curves <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040175.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040176.png" />, of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040177.png" />. Each of the curves <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040178.png" /> has <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040179.png" /> double points and passes through <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040180.png" /> pinch points of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040181.png" />. The octic is a <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040182.png" /> covering of a [[Kummer surface|Kummer surface]]: <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040183.png" /> (see also Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040184.png" /> below). The restrictions <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040185.png" /> are <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040186.png" />-fold coverings of four double conics of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040187.png" /> lying on a coordinate tetrahedron. The three double points of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040188.png" /> map to three double points of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040189.png" /> on the conic <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040190.png" /> and the <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040191.png" /> pinch points on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040192.png" /> map to the other three double points on the double conic <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040193.png" /> (see [[#References|[a6]]]). | + | Type $ ( 1,4 ) $— |
| + | There are $ 24 $ |
| + | isogenies $ f : {( A, \lambda ) } \rightarrow {( B, \theta ) } $ |
| + | of degree $ 4 $ |
| + | onto principally polarized Abelian surfaces. If the curves $ C $ |
| + | and $ H $ |
| + | do not admit elliptic involutions compatible with $ f $, |
| + | then $ \varphi _ {L} :A \twoheadrightarrow {\overline{A}\; } \subset \mathbf P ^ {3} $ |
| + | is a birational morphism onto a singular octic $ {\overline{A}\; } $. |
| + | In the exceptional case, $ \varphi _ {L} : A \twoheadrightarrow {\overline{A}\; } \subset \mathbf P ^ {3} $ |
| + | is a double covering of a singular quartic $ {\overline{A}\; } $, |
| + | which is birational to an elliptic scroll. In the first case the octic $ {\overline{A}\; } $ |
| + | is smooth outside the four coordinate planes of $ \mathbf P ^ {3} $ |
| + | and touches the coordinate planes in curves $ D _ {i} $, |
| + | $ i = 1 \dots 4 $, |
| + | of degree $ 4 $. |
| + | Each of the curves $ D _ {i} $ |
| + | has $ 3 $ |
| + | double points and passes through $ 12 $ |
| + | pinch points of $ {\overline{A}\; } $. |
| + | The octic is a $ 8:1 $ |
| + | covering of a [[Kummer surface|Kummer surface]]: $ p: {\overline{A}\; } \twoheadrightarrow K \subset \mathbf P ^ {3} $( |
| + | see also Type $ ( 2,2 ) $ |
| + | below). The restrictions $ p \mid _ {D _ {i} } $ |
| + | are $ 4 $- |
| + | fold coverings of four double conics of $ K $ |
| + | lying on a coordinate tetrahedron. The three double points of $ D _ {i} $ |
| + | map to three double points of $ K $ |
| + | on the conic $ p ( D _ {i} ) $ |
| + | and the $ 12 $ |
| + | pinch points on $ D _ {i} $ |
| + | map to the other three double points on the double conic $ p ( D _ {i} ) $( |
| + | see [[#References|[a6]]]). |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040194.png" />— The invertible sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040195.png" /> is very ample, i.e. <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040196.png" /> is an embedding if and only if the curves <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040197.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040198.png" /> do not admit elliptic involutions compatible with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040199.png" />. In the exceptional case <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040200.png" /> is a double covering of an elliptic scroll (see [[#References|[a13]]] and [[#References|[a9]]]). If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040201.png" /> is very ample, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040202.png" /> is a smooth surface of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040203.png" /> in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040204.png" />. It is the zero locus of a section of the Horrocks–Mumford bundle <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040205.png" /> (see [[#References|[a8]]]). Conversely, the zero set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040206.png" /> of a general section <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040207.png" /> is an Abelian surface of degree <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040208.png" />, i.e. of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040209.png" />. | + | Type $ ( 1,5 ) $— |
| + | The invertible sheaf $ L $ |
| + | is very ample, i.e. $ {\varphi _ {L} } : A \rightarrow {\mathbf P ^ {4} } $ |
| + | is an embedding if and only if the curves $ C $ |
| + | and $ H $ |
| + | do not admit elliptic involutions compatible with $ f $. |
| + | In the exceptional case $ \varphi _ {L} $ |
| + | is a double covering of an elliptic scroll (see [[#References|[a13]]] and [[#References|[a9]]]). If $ L $ |
| + | is very ample, $ \varphi _ {L} ( A ) $ |
| + | is a smooth surface of degree $ 10 $ |
| + | in $ \mathbf P ^ {4} $. |
| + | It is the zero locus of a section of the Horrocks–Mumford bundle $ F $( |
| + | see [[#References|[a8]]]). Conversely, the zero set $ \{ \sigma = 0 \} \subset \mathbf P ^ {4} $ |
| + | of a general section $ \sigma \in H ^ {0} ( \mathbf P ^ {4} ,F ) $ |
| + | is an Abelian surface of degree $ 10 $, |
| + | i.e. of type $ ( 1,5 ) $. |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040210.png" />— <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040211.png" /> is twice a principal polarization on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040212.png" />. The morphism <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040213.png" /> is a double covering of the [[Kummer surface|Kummer surface]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040214.png" /> associated with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040215.png" />. It is isomorphic to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040216.png" />. | + | Type $ ( 2,2 ) $— |
| + | $ \lambda $ |
| + | is twice a principal polarization on $ A $. |
| + | The morphism $ \varphi _ {L} : A \twoheadrightarrow K _ {A} \subset \mathbf P ^ {3} $ |
| + | is a double covering of the [[Kummer surface|Kummer surface]] $ K _ {A} $ |
| + | associated with $ A $. |
| + | It is isomorphic to $ {A / {( - 1 ) _ {A} } } $. |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040217.png" />— The ideal sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040218.png" /> of the image of the embedding <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040219.png" /> is generated by <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040220.png" /> quadrics (see [[#References|[a3]]]). | + | Type $ ( 2,4 ) $— |
| + | The ideal sheaf $ {\mathcal I} _ { {A / {\mathbf P ^ {7} } } } $ |
| + | of the image of the embedding $ \varphi _ {L} : A \hookrightarrow \mathbf P ^ {7} $ |
| + | is generated by $ 6 $ |
| + | quadrics (see [[#References|[a3]]]). |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040221.png" />— Suppose <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040222.png" /> is very ample and let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040223.png" /> be the associated Kummer surface. The subvector space <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040224.png" /> of odd sections induces an embedding of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040225.png" />, the blow-up of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040226.png" /> in the <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040227.png" /> double points, as a smooth quartic surface into <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040228.png" />. <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040229.png" /> is invariant under the action of the level-<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040230.png" /> Heisenberg group (cf. also [[Heisenberg representation|Heisenberg representation]]) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040231.png" />. The <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040232.png" /> blown-up double points become skew lines on the quartic surface. Any <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040233.png" />-invariant quartic surface in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040234.png" /> with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040235.png" /> skew lines comes from a polarized Abelian surface <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040236.png" /> of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040237.png" /> in this way (see [[#References|[a5]]], [[#References|[a11]]] and [[#References|[a12]]]). | + | Type $ ( 2,6 ) $— |
| + | Suppose $ L $ |
| + | is very ample and let $ K _ {A} = {A / {( - 1 ) _ {A} } } $ |
| + | be the associated Kummer surface. The subvector space $ H ^ {0} ( A,L ) ^ {-} \subset H ^ {0} ( A,L ) $ |
| + | of odd sections induces an embedding of $ {\widetilde{K} } _ {A} $, |
| + | the blow-up of $ K _ {A} $ |
| + | in the $ 16 $ |
| + | double points, as a smooth quartic surface into $ \mathbf P ^ {3} $. |
| + | $ {\widetilde{K} } _ {A} \subset \mathbf P ^ {3} $ |
| + | is invariant under the action of the level- $ 2 $ |
| + | Heisenberg group (cf. also [[Heisenberg representation|Heisenberg representation]]) $ H ( 2,2 ) $. |
| + | The $ 16 $ |
| + | blown-up double points become skew lines on the quartic surface. Any $ H ( 2,2 ) $- |
| + | invariant quartic surface in $ \mathbf P ^ {3} $ |
| + | with $ 16 $ |
| + | skew lines comes from a polarized Abelian surface $ ( A, \lambda ) $ |
| + | of type $ ( 2,6 ) $ |
| + | in this way (see [[#References|[a5]]], [[#References|[a11]]] and [[#References|[a12]]]). |
| | | |
− | Type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040238.png" />— <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040239.png" /> is three times a principal polarization and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040240.png" /> is an embedding. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040241.png" /> is not a product, then the quadrics <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040242.png" /> vanishing on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040243.png" /> generate the ideal sheaf <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040244.png" />. In the product case, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040245.png" /> is generated by quadrics and cubics (see [[#References|[a4]]]). | + | Type $ ( 3,3 ) $— |
| + | $ \lambda $ |
| + | is three times a principal polarization and $ \varphi _ {L} : A \hookrightarrow \mathbf P ^ {8} $ |
| + | is an embedding. If $ ( A, \lambda ) $ |
| + | is not a product, then the quadrics $ Q \in H ^ {0} ( \mathbf P ^ {8} , {\mathcal I} _ { {A / {\mathbf P ^ {8} } } } ( 2 ) ) $ |
| + | vanishing on $ A $ |
| + | generate the ideal sheaf $ {\mathcal I} _ { {A / {\mathbf P ^ {8} } } } $. |
| + | In the product case, $ {\mathcal I} _ { {A / {\mathbf P ^ {8} } } } $ |
| + | is generated by quadrics and cubics (see [[#References|[a4]]]). |
| | | |
| ==Algebraic completely integrable systems.== | | ==Algebraic completely integrable systems.== |
− | An algebraic completely integrable system in the sense of M. Adler and P. van Moerbeke is a completely integrable polynomial [[Hamiltonian system|Hamiltonian system]] on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040246.png" /> (with Casimir functions <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040247.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040248.png" /> independent constants of motion <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040249.png" /> in involution) such that: | + | An algebraic completely integrable system in the sense of M. Adler and P. van Moerbeke is a completely integrable polynomial [[Hamiltonian system|Hamiltonian system]] on $ \mathbf C ^ {N} $( |
| + | with Casimir functions $ {H _ {1} \dots H _ {k} } : {\mathbf C ^ {N} } \rightarrow \mathbf C $ |
| + | and $ m = { {( N - k ) } / 2 } $ |
| + | independent constants of motion $ H _ {k + 1 } \dots H _ {k + m } $ |
| + | in involution) such that: |
| | | |
− | a) for a general point <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040250.png" /> the invariant manifold <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040251.png" /> is an open affine part of an Abelian variety <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040252.png" />; | + | a) for a general point $ c = {} ^ {t} ( c _ {1} \dots c _ {k + m } ) \in \mathbf C ^ {k + m } $ |
| + | the invariant manifold $ A _ {c} ^ {o} = \cap _ {i = 1 } ^ {m + k } \{ H _ {i} = c _ {i} \} \subset \mathbf C ^ {N} $ |
| + | is an open affine part of an Abelian variety $ A _ {c} $; |
| | | |
− | b) the flows of the integrable vector fields <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040253.png" /> linearize on the Abelian varieties <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040254.png" /> [[#References|[a2]]]. | + | b) the flows of the integrable vector fields $ X _ {u _ {i} } $ |
| + | linearize on the Abelian varieties $ A _ {c} $[[#References|[a2]]]. |
| | | |
− | The divisor at infinity <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040255.png" /> defines a polarization on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040256.png" />. In this way the mapping <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040257.png" /> defines a family of polarized Abelian varieties (cf. [[Moduli problem|Moduli problem]]). Some examples of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040258.png" />-dimensional algebraic completely integrable systems and their associated Abelian surfaces are: | + | The divisor at infinity $ D _ {c} = A _ {c} - A _ {c} ^ {o} $ |
| + | defines a polarization on $ A _ {c} $. |
| + | In this way the mapping $ {( H _ {1} \dots H _ {k + m } ) } : {\mathbf C ^ {N} } \rightarrow {\mathbf C ^ {k + m } } $ |
| + | defines a family of polarized Abelian varieties (cf. [[Moduli problem|Moduli problem]]). Some examples of $ 2 $- |
| + | dimensional algebraic completely integrable systems and their associated Abelian surfaces are: |
| | | |
| the three-body Toda lattice and the even, respectively odd, master systems (cf. also [[Master equations in cooperative and social phenomena|Master equations in cooperative and social phenomena]]) linearize on principally polarized Abelian surfaces; | | the three-body Toda lattice and the even, respectively odd, master systems (cf. also [[Master equations in cooperative and social phenomena|Master equations in cooperative and social phenomena]]) linearize on principally polarized Abelian surfaces; |
| | | |
− | the [[Kowalewski top|Kowalewski top]], the Hénon–Heiles system and the Manakov geodesic flow on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040259.png" /> linearize on Abelian surfaces of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040260.png" /> [[#References|[a1]]]; | + | the [[Kowalewski top|Kowalewski top]], the Hénon–Heiles system and the Manakov geodesic flow on $ { \mathop{\rm SO} } ( 4 ) $ |
| + | linearize on Abelian surfaces of type $ ( 1,2 ) $[[#References|[a1]]]; |
| | | |
− | the Garnier system linearizes on Abelian surfaces of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040261.png" /> [[#References|[a15]]]. | + | the Garnier system linearizes on Abelian surfaces of type $ ( 1,4 ) $[[#References|[a15]]]. |
| | | |
| ====References==== | | ====References==== |
− | <table><TR><TD valign="top">[a1]</TD> <TD valign="top"> M. Adler, P. van Moerbeke, "The Kowalewski and Hénon–Heiles motions as Manakov geodesic flows on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040262.png" />: a two-dimensional family of Lax pairs" ''Comm. Math. Phys.'' , '''113''' (1988) pp. 659–700</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top"> M. Adler, P. van Moerbeke, "The complex geometry of the Kowalewski–Painlevé analysis" ''Invent. Math.'' , '''97''' (1989) pp. 3–51 {{MR|}} {{ZBL|0678.58020}} </TD></TR><TR><TD valign="top">[a3]</TD> <TD valign="top"> W. Barth, "Abelian surfaces with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040263.png" />-polarization" , ''Algebraic Geometry, Sendai, 1985'' , ''Advanced Studies in Pure Math.'' , '''10''' (1987) pp. 41–84 {{MR|946234}} {{ZBL|}} </TD></TR><TR><TD valign="top">[a4]</TD> <TD valign="top"> W. Barth, "Quadratic equations for level-<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040264.png" /> abelian surfaces" , ''Abelian Varieties, Proc. Workshop Egloffstein 1993'' , de Gruyter (1995) pp. 1–18 {{MR|1336597}} {{ZBL|}} </TD></TR><TR><TD valign="top">[a5]</TD> <TD valign="top"> W. Barth, I. Nieto, "Abelian surfaces of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040265.png" /> and quartic surfaces with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040266.png" /> skew lines" ''J. Algebraic Geom.'' , '''3''' (1994) pp. 173–222 {{MR|1257320}} {{ZBL|0809.14027}} </TD></TR><TR><TD valign="top">[a6]</TD> <TD valign="top"> Ch. Birkenhake, H. Lange, D. van Straten, "Abelian surfaces of type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040267.png" />" ''Math. Ann.'' , '''285''' (1989) pp. 625–646 {{MR|1027763}} {{ZBL|0714.14028}} </TD></TR><TR><TD valign="top">[a7]</TD> <TD valign="top"> Ch. Birkenhake, H. Lange, "Moduli spaces of Abelian surfaces wih isogeny" , ''Geometry and Analysis, Bombay Colloquium 1992'' , Tata Inst. Fundam. Res. (1995) pp. 225–243 {{MR|1351509}} {{ZBL|}} </TD></TR><TR><TD valign="top">[a8]</TD> <TD valign="top"> G. Horrocks, D. Mumford, "A rank <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040268.png" /> vector bundle on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040269.png" /> with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040270.png" /> symmetries" ''Topology'' , '''12''' (1973) pp. 63–81 {{MR|382279}} {{ZBL|0255.14017}} </TD></TR><TR><TD valign="top">[a9]</TD> <TD valign="top"> K. Hulek, H. Lange, "Examples of abelian surfaces in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040271.png" />" ''J. Reine Angew. Math.'' , '''363''' (1985) pp. 200–216 {{MR|0814021}} {{ZBL|0593.14027}} </TD></TR><TR><TD valign="top">[a10]</TD> <TD valign="top"> H. Lange, Ch. Birkenhake, "Complex Abelian varieties" , ''Grundlehren math. Wiss.'' , '''302''' , Springer (1992) {{MR|1217487}} {{ZBL|0779.14012}} </TD></TR><TR><TD valign="top">[a11]</TD> <TD valign="top"> I. Naruki, "On smooth quartic embeddings of Kummer surfaces" ''Proc. Japan Acad.'' , '''67 A''' (1991) pp. 223–224 {{MR|1137912}} {{ZBL|}} </TD></TR><TR><TD valign="top">[a12]</TD> <TD valign="top"> V. V. Nikulin, "On Kummer surfaces" ''Math USSR Izv.'' , '''9''' (1975) pp. 261–275 (In Russian) {{MR|429917}} {{ZBL|0325.14015}} </TD></TR><TR><TD valign="top">[a13]</TD> <TD valign="top"> S. Ramanan, "Ample divisors on abelian surfaces" ''Proc. London Math. Soc.'' , '''51''' (1985) pp. 231–245 {{MR|0794112}} {{ZBL|0603.14013}} </TD></TR><TR><TD valign="top">[a14]</TD> <TD valign="top"> I. Reider, "Vector bundles of rank <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040272.png" /> and linear systems on algebraic surfaces" ''Ann. of Math.'' , '''127''' (1988) pp. 309–316 {{MR|0932299}} {{ZBL|0663.14010}} </TD></TR><TR><TD valign="top">[a15]</TD> <TD valign="top"> P. Vanhaecke, "A special case of the Garnier system, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/a/a110/a110040/a110040273.png" />-polarized Abelian surfaces and their moduli" ''Compositio Math.'' , '''92''' (1994) pp. 157–203 {{MR|1283227}} {{ZBL|}} </TD></TR></table> | + | <table> |
| + | <TR><TD valign="top">[a1]</TD> <TD valign="top"> M. Adler, P. van Moerbeke, "The Kowalewski and Hénon–Heiles motions as Manakov geodesic flows on $SO(4)$: a two-dimensional family of Lax pairs" ''Comm. Math. Phys.'' , '''113''' (1988) pp. 659–700</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top"> M. Adler, P. van Moerbeke, "The complex geometry of the Kowalewski–Painlevé analysis" ''Invent. Math.'' , '''97''' (1989) pp. 3–51 {{MR|}} {{ZBL|0678.58020}} </TD></TR> |
| + | <TR><TD valign="top">[a3]</TD> <TD valign="top"> W. Barth, "Abelian surfaces with $(1,2)$-polarization" , ''Algebraic Geometry, Sendai, 1985'' , ''Advanced Studies in Pure Math.'' , '''10''' (1987) pp. 41–84 {{MR|946234}} {{ZBL|}} </TD></TR> |
| + | <TR><TD valign="top">[a4]</TD> <TD valign="top"> W. Barth, "Quadratic equations for level-$3$ abelian surfaces" , ''Abelian Varieties, Proc. Workshop Egloffstein 1993'' , de Gruyter (1995) pp. 1–18 {{MR|1336597}} {{ZBL|}} </TD></TR> |
| + | <TR><TD valign="top">[a5]</TD> <TD valign="top"> W. Barth, I. Nieto, "Abelian surfaces of type $(1,3)$ and quartic surfaces with $16$ skew lines" ''J. Algebraic Geom.'' , '''3''' (1994) pp. 173–222 {{MR|1257320}} {{ZBL|0809.14027}} </TD></TR> |
| + | <TR><TD valign="top">[a6]</TD> <TD valign="top"> Ch. Birkenhake, H. Lange, D. van Straten, "Abelian surfaces of type $(1,4)$" ''Math. Ann.'' , '''285''' (1989) pp. 625–646 {{MR|1027763}} {{ZBL|0714.14028}} </TD></TR> |
| + | <TR><TD valign="top">[a7]</TD> <TD valign="top"> Ch. Birkenhake, H. Lange, "Moduli spaces of Abelian surfaces wih isogeny" , ''Geometry and Analysis, Bombay Colloquium 1992'' , Tata Inst. Fundam. Res. (1995) pp. 225–243 {{MR|1351509}} {{ZBL|}} </TD></TR> |
| + | <TR><TD valign="top">[a8]</TD> <TD valign="top"> G. Horrocks, D. Mumford, "A rank $2$ vector bundle on $\mathbb{P}^4$ with $15000$ symmetries" ''Topology'' , '''12''' (1973) pp. 63–81 {{MR|382279}} {{ZBL|0255.14017}} </TD></TR> |
| + | <TR><TD valign="top">[a9]</TD> <TD valign="top"> K. Hulek, H. Lange, "Examples of abelian surfaces in $\mathbb{P}^4$" ''J. Reine Angew. Math.'' , '''363''' (1985) pp. 200–216 {{MR|0814021}} {{ZBL|0593.14027}} </TD></TR> |
| + | <TR><TD valign="top">[a10]</TD> <TD valign="top"> H. Lange, Ch. Birkenhake, "Complex Abelian varieties" , ''Grundlehren math. Wiss.'' , '''302''' , Springer (1992) {{MR|1217487}} {{ZBL|0779.14012}} </TD></TR><TR><TD valign="top">[a11]</TD> <TD valign="top"> I. Naruki, "On smooth quartic embeddings of Kummer surfaces" ''Proc. Japan Acad.'' , '''67 A''' (1991) pp. 223–224 {{MR|1137912}} {{ZBL|}} </TD></TR><TR><TD valign="top">[a12]</TD> <TD valign="top"> V. V. Nikulin, "On Kummer surfaces" ''Math USSR Izv.'' , '''9''' (1975) pp. 261–275 (In Russian) {{MR|429917}} {{ZBL|0325.14015}} </TD></TR> |
| + | <TR><TD valign="top">[a13]</TD> <TD valign="top"> S. Ramanan, "Ample divisors on abelian surfaces" ''Proc. London Math. Soc.'' , '''51''' (1985) pp. 231–245 {{MR|0794112}} {{ZBL|0603.14013}} </TD></TR> |
| + | <TR><TD valign="top">[a14]</TD> <TD valign="top"> I. Reider, "Vector bundles of rank $2$ and linear systems on algebraic surfaces" ''Ann. of Math.'' , '''127''' (1988) pp. 309–316 {{MR|0932299}} {{ZBL|0663.14010}} </TD></TR><TR><TD valign="top">[a15]</TD> <TD valign="top"> P. Vanhaecke, "A special case of the Garnier system, $(1,4)$-polarized Abelian surfaces and their moduli" ''Compositio Math.'' , '''92''' (1994) pp. 157–203 {{MR|1283227}} {{ZBL|}} </TD></TR></table> |
An Abelian variety of dimension two, i.e. a complete connected group variety of dimension two over a field $ k $.
The group law is commutative. In the sequel, $ k $
is assumed to be algebraically closed (cf. Algebraically closed field).
In the classification of algebraic surfaces, Abelian surfaces are exactly the smooth complete surfaces $ A $
with Kodaira dimension $ \kappa = 0 $,
geometric genus $ p _ {g} = h ^ {2} ( A, {\mathcal O} _ {A} ) =1 $
and irregularity $ q = h ^ {1} ( A, {\mathcal O} _ {A} ) = 2 $.
For an Abelian surface $ A $,
the dual Abelian variety $ {\widehat{A} } = { \mathop{\rm Pic} } ^ {0} ( A ) $
is again an Abelian surface. An invertible sheaf $ L $
on $ A $
defines the homomorphism $ {\phi _ {L} } : A \rightarrow { {\widehat{A} } } $,
$ a \mapsto t _ {a} ^ {*} L \otimes L ^ {- 1 } $.
The homomorphism $ \phi _ {L} $
depends only on the algebraic equivalence class of $ L $.
The invertible sheaf $ L $
is ample (cf. Ample sheaf) if and only if $ \phi _ {L} $
is an isogeny (i.e., $ \phi _ {L} $
is surjective and has finite kernel) and $ h ^ {0} ( A,L ) \neq0 $.
In this case, $ { \mathop{\rm deg} } \phi _ {L} = d ^ {2} $
with a positive integer $ d $
and the Riemann–Roch theorem says that
$$
h ^ {0} ( A,L ) = {
\frac{1}{2}
} ( L ^ {2} ) = d,
$$
where $ ( L ^ {2} ) $
denotes the self-intersection number. Every Abelian surface admits an ample invertible sheaf and hence is projective (cf. Projective scheme).
A polarization $ \lambda $
on $ A $
is the algebraic equivalence class $ [ L ] $
of an ample invertible sheaf $ L $.
The degree $ { \mathop{\rm deg} } \lambda $
of $ \lambda $
is by definition $ d = \sqrt { { \mathop{\rm deg} } \phi _ {L} } $.
An Abelian surface $ A $
together with a polarization $ \lambda $
is a polarized Abelian surface. A principal polarization is a polarization of degree $ 1 $.
A principally polarized Abelian surface $ ( A, \lambda ) $
is either the Jacobi variety $ J ( H ) $
of a smooth projective curve $ H $
of genus $ 2 $,
and $ \lambda = \theta $
is the class of the theta divisor, or $ A $
is the product of two elliptic curves (Abelian varieties of dimension one, cf. also Elliptic curve) with $ \lambda $
the product polarization.
If the degree of $ \lambda = [ L ] $
is prime to $ { \mathop{\rm char} } ( k ) $,
then $ \lambda $
is said to be a separable polarization and the kernel of $ \phi _ {L} $
is isomorphic to the group $ ( \mathbf Z/d _ {1} \mathbf Z ) ^ {2} \times ( \mathbf Z/d _ {2} \mathbf Z ) ^ {2} $,
where $ d _ {1} $
and $ d _ {2} $
are positive integers such that $ d _ {1} $
divides $ d _ {2} $
and $ d _ {1} d _ {2} = { \mathop{\rm deg} } \lambda $.
The pair $ ( d _ {1} ,d _ {2} ) $
is the type of the polarized Abelian surface $ ( A, \lambda ) $.
A polarization $ \lambda = [ L ] $
of type $ ( d _ {1} ,d _ {2} ) $
on $ A $
defines a polarization $ {\widehat \lambda } = [ {\widehat{L} } ] $
on the dual Abelian surface $ {\widehat{A} } $.
The polarization $ {\widehat \lambda } $
is again of type $ ( d _ {1} ,d _ {2} ) $
and it is characterized by each of the following two equivalent properties:
$$
\phi _ {L} ^ {*} {\widehat \lambda } = d _ {1} d _ {2} \lambda \iff \phi _ { {\widehat{L} } } \phi _ {L} = d _ {1} d _ {2} { \mathop{\rm id} } _ {A} .
$$
For a polarized Abelian surface $ ( A, \lambda = [ L ] ) $
of type $ ( d _ {1} ,d _ {2} ) $,
the assignment $ A \ni a \mapsto \{ {\sigma \in H ^ {0} ( A,L ) } : {\sigma ( a ) = 0 } \} \subset H ^ {0} ( A,L ) $
defines a rational mapping from $ A $
into the projective space of hyperplanes in $ H ^ {0} ( A,L ) $:
$$
{\varphi _ {L} } : A \rightarrow {\mathbf P ( H ^ {0} ( A,L ) ^ {*} ) } \simeq \mathbf P _ {k} ^ {d _ {1} d _ {2} - 1 } .
$$
If $ d _ {1} \geq 2 $,
then $ \varphi _ {L} $
is everywhere defined. The Lefschetz theorem says that for $ d _ {1} \geq 3 $
the morphism $ \varphi _ {L} $
is an embedding. Suppose $ d _ {1} = 2 $;
then $ \lambda = 2 \mu $
with a polarization $ \mu = [ M ] $
of type $ ( 1, { {d _ {2} } / 2 } ) $.
If the linear system $ | M | $
has no fixed components, then $ \varphi _ {L} $
is an embedding.
Complex Abelian surfaces.
An Abelian surface over the field $ \mathbf C $
of complex numbers is a complex torus
$$
A = { {\mathbf C ^ {2} } / \Lambda }
$$
(with a lattice $ \Lambda \simeq \mathbf Z ^ {4} $
in $ \mathbf C ^ {2} $)
admitting a polarization. A polarization $ \lambda $
on $ A $
can be considered as a non-degenerate alternating form $ \Lambda \times \Lambda \rightarrow \mathbf Z $,
the elementary divisors of which are given by the type $ ( d _ {1} ,d _ {2} ) $
of $ \lambda $.
In the sequel, the field $ k $
is assumed to be $ \mathbf C $,
although some of the following results are also valid for arbitrary algebraically closed fields.
Suppose $ ( A, \lambda = [ L ] ) $
is of type $ ( 1,d ) $
and the linear system $ | L | $
has no fixed components. The Reider theorem states that for $ d \geq 5 $
the invertible sheaf $ L $
is very ample if and only if there is no elliptic curve $ E $
on $ A $
with $ ( E \cdot L ) = 2 $(
see [a14] and [a10]). For arbitrary $ d \geq 1 $
there exist finitely many isogenies $ f : {( A, \lambda ) } \rightarrow {( B, \theta ) } $
of degree $ d $
onto principally polarized Abelian surfaces (cf. also Isogeny). Suppose $ \theta = [ \Theta ] $
with a symmetric invertible sheaf $ \Theta $(
i.e., $ ( -1 ) _ {A} ^ {*} \Theta \simeq \Theta $)
and let $ H $
be the unique divisor in the linear system $ | \Theta | $.
The divisor $ C = f ^ {- 1 } ( H ) $
on $ A $
defines a symmetric invertible sheaf $ L = {\mathcal O} _ {A} ( C ) $
with class $ [ L ] = \lambda $
and the covering $ {f \mid _ {C} } : C \rightarrow H $
is étale of degree $ d $.
One distinguishes two cases: i) $ H $
is smooth of genus $ 2 $,
$ B = J ( H ) $
and $ C $
is smooth of genus $ d + 1 $;
and ii) $ H $
is the sum $ E _ {1} + E _ {2} $
of two elliptic curves with intersection number $ ( E _ {1} \cdot E _ {2} ) = 1 $,
$ B = E _ {1} \times E _ {2} $
and $ C $
is the sum $ F _ {1} + F _ {2} $
of two elliptic curves with $ ( F _ {1} \cdot F _ {2} ) = d $.
In the following list, $ ( A, \lambda = [ L ] ) $
is a polarized Abelian surface of type $ ( d _ {1} ,d _ {2} ) $
such that $ | L | $
admits no fixed components
Type $ ( 1,2 ) $—
The linear system $ | L | $
has exactly $ 4 $
base points. The blow-up $ {\widetilde{A} } $
of $ A $
in these points admits a morphism $ { {\widetilde \varphi } _ {L} } : { {\widetilde{A} } } \rightarrow {\mathbf P ^ {1} } $
induced by $ \varphi _ {L} $.
The general fibre of $ {\widetilde \varphi } _ {L} $
is a smooth curve of genus $ 3 $.
The curve $ C $
on $ A $
defining $ L $
as above is double elliptic: $ C { \mathop \rightarrow \limits ^ { {2:1 }} } E $
with an elliptic curve $ E $,
and $ A $
is isomorphic to $ { {J ( C ) } / E } $(
see [a3]).
Type $ ( 1,3 ) $—
$ L $
defines a $ 6 $-
fold covering $ {\varphi _ {L} } : A \rightarrow {\mathbf P ^ {2} } $
ramified along a curve $ R \subset \mathbf P ^ {2} $
of degree $ 18 $.
The general divisor in the linear system $ | L | $
is a smooth curve of genus $ 4 $.
There are $ 4 $
isogenies $ f : {( A, \lambda ) } \rightarrow {( B, \theta ) } $
of degree $ 3 $
onto principally polarized Abelian surfaces. In case i) the smooth genus- $ 4 $
curve $ C \in | L | $
is double elliptic: $ C { \mathop \rightarrow \limits ^ { {2:1 }} } E $,
and the embedding of $ E $
into the Jacobian $ J ( C ) $
induces an exact sequence
$$
0 \rightarrow E \times E \rightarrow J ( C ) \rightarrow A \rightarrow 0.
$$
The étale $ 3 $-
fold covering $ {f \mid _ {C} } : C \rightarrow H $
induces a morphism $ {f ^ {*} } : {J ( H ) } \rightarrow {J ( C ) } $
with image $ {\widehat{A} } $,
the dual Abelian surface of $ A $(
see [a7]).
Type $ ( 1,4 ) $—
There are $ 24 $
isogenies $ f : {( A, \lambda ) } \rightarrow {( B, \theta ) } $
of degree $ 4 $
onto principally polarized Abelian surfaces. If the curves $ C $
and $ H $
do not admit elliptic involutions compatible with $ f $,
then $ \varphi _ {L} :A \twoheadrightarrow {\overline{A}\; } \subset \mathbf P ^ {3} $
is a birational morphism onto a singular octic $ {\overline{A}\; } $.
In the exceptional case, $ \varphi _ {L} : A \twoheadrightarrow {\overline{A}\; } \subset \mathbf P ^ {3} $
is a double covering of a singular quartic $ {\overline{A}\; } $,
which is birational to an elliptic scroll. In the first case the octic $ {\overline{A}\; } $
is smooth outside the four coordinate planes of $ \mathbf P ^ {3} $
and touches the coordinate planes in curves $ D _ {i} $,
$ i = 1 \dots 4 $,
of degree $ 4 $.
Each of the curves $ D _ {i} $
has $ 3 $
double points and passes through $ 12 $
pinch points of $ {\overline{A}\; } $.
The octic is a $ 8:1 $
covering of a Kummer surface: $ p: {\overline{A}\; } \twoheadrightarrow K \subset \mathbf P ^ {3} $(
see also Type $ ( 2,2 ) $
below). The restrictions $ p \mid _ {D _ {i} } $
are $ 4 $-
fold coverings of four double conics of $ K $
lying on a coordinate tetrahedron. The three double points of $ D _ {i} $
map to three double points of $ K $
on the conic $ p ( D _ {i} ) $
and the $ 12 $
pinch points on $ D _ {i} $
map to the other three double points on the double conic $ p ( D _ {i} ) $(
see [a6]).
Type $ ( 1,5 ) $—
The invertible sheaf $ L $
is very ample, i.e. $ {\varphi _ {L} } : A \rightarrow {\mathbf P ^ {4} } $
is an embedding if and only if the curves $ C $
and $ H $
do not admit elliptic involutions compatible with $ f $.
In the exceptional case $ \varphi _ {L} $
is a double covering of an elliptic scroll (see [a13] and [a9]). If $ L $
is very ample, $ \varphi _ {L} ( A ) $
is a smooth surface of degree $ 10 $
in $ \mathbf P ^ {4} $.
It is the zero locus of a section of the Horrocks–Mumford bundle $ F $(
see [a8]). Conversely, the zero set $ \{ \sigma = 0 \} \subset \mathbf P ^ {4} $
of a general section $ \sigma \in H ^ {0} ( \mathbf P ^ {4} ,F ) $
is an Abelian surface of degree $ 10 $,
i.e. of type $ ( 1,5 ) $.
Type $ ( 2,2 ) $—
$ \lambda $
is twice a principal polarization on $ A $.
The morphism $ \varphi _ {L} : A \twoheadrightarrow K _ {A} \subset \mathbf P ^ {3} $
is a double covering of the Kummer surface $ K _ {A} $
associated with $ A $.
It is isomorphic to $ {A / {( - 1 ) _ {A} } } $.
Type $ ( 2,4 ) $—
The ideal sheaf $ {\mathcal I} _ { {A / {\mathbf P ^ {7} } } } $
of the image of the embedding $ \varphi _ {L} : A \hookrightarrow \mathbf P ^ {7} $
is generated by $ 6 $
quadrics (see [a3]).
Type $ ( 2,6 ) $—
Suppose $ L $
is very ample and let $ K _ {A} = {A / {( - 1 ) _ {A} } } $
be the associated Kummer surface. The subvector space $ H ^ {0} ( A,L ) ^ {-} \subset H ^ {0} ( A,L ) $
of odd sections induces an embedding of $ {\widetilde{K} } _ {A} $,
the blow-up of $ K _ {A} $
in the $ 16 $
double points, as a smooth quartic surface into $ \mathbf P ^ {3} $.
$ {\widetilde{K} } _ {A} \subset \mathbf P ^ {3} $
is invariant under the action of the level- $ 2 $
Heisenberg group (cf. also Heisenberg representation) $ H ( 2,2 ) $.
The $ 16 $
blown-up double points become skew lines on the quartic surface. Any $ H ( 2,2 ) $-
invariant quartic surface in $ \mathbf P ^ {3} $
with $ 16 $
skew lines comes from a polarized Abelian surface $ ( A, \lambda ) $
of type $ ( 2,6 ) $
in this way (see [a5], [a11] and [a12]).
Type $ ( 3,3 ) $—
$ \lambda $
is three times a principal polarization and $ \varphi _ {L} : A \hookrightarrow \mathbf P ^ {8} $
is an embedding. If $ ( A, \lambda ) $
is not a product, then the quadrics $ Q \in H ^ {0} ( \mathbf P ^ {8} , {\mathcal I} _ { {A / {\mathbf P ^ {8} } } } ( 2 ) ) $
vanishing on $ A $
generate the ideal sheaf $ {\mathcal I} _ { {A / {\mathbf P ^ {8} } } } $.
In the product case, $ {\mathcal I} _ { {A / {\mathbf P ^ {8} } } } $
is generated by quadrics and cubics (see [a4]).
Algebraic completely integrable systems.
An algebraic completely integrable system in the sense of M. Adler and P. van Moerbeke is a completely integrable polynomial Hamiltonian system on $ \mathbf C ^ {N} $(
with Casimir functions $ {H _ {1} \dots H _ {k} } : {\mathbf C ^ {N} } \rightarrow \mathbf C $
and $ m = { {( N - k ) } / 2 } $
independent constants of motion $ H _ {k + 1 } \dots H _ {k + m } $
in involution) such that:
a) for a general point $ c = {} ^ {t} ( c _ {1} \dots c _ {k + m } ) \in \mathbf C ^ {k + m } $
the invariant manifold $ A _ {c} ^ {o} = \cap _ {i = 1 } ^ {m + k } \{ H _ {i} = c _ {i} \} \subset \mathbf C ^ {N} $
is an open affine part of an Abelian variety $ A _ {c} $;
b) the flows of the integrable vector fields $ X _ {u _ {i} } $
linearize on the Abelian varieties $ A _ {c} $[a2].
The divisor at infinity $ D _ {c} = A _ {c} - A _ {c} ^ {o} $
defines a polarization on $ A _ {c} $.
In this way the mapping $ {( H _ {1} \dots H _ {k + m } ) } : {\mathbf C ^ {N} } \rightarrow {\mathbf C ^ {k + m } } $
defines a family of polarized Abelian varieties (cf. Moduli problem). Some examples of $ 2 $-
dimensional algebraic completely integrable systems and their associated Abelian surfaces are:
the three-body Toda lattice and the even, respectively odd, master systems (cf. also Master equations in cooperative and social phenomena) linearize on principally polarized Abelian surfaces;
the Kowalewski top, the Hénon–Heiles system and the Manakov geodesic flow on $ { \mathop{\rm SO} } ( 4 ) $
linearize on Abelian surfaces of type $ ( 1,2 ) $[a1];
the Garnier system linearizes on Abelian surfaces of type $ ( 1,4 ) $[a15].
References
[a1] | M. Adler, P. van Moerbeke, "The Kowalewski and Hénon–Heiles motions as Manakov geodesic flows on $SO(4)$: a two-dimensional family of Lax pairs" Comm. Math. Phys. , 113 (1988) pp. 659–700 |
[a2] | M. Adler, P. van Moerbeke, "The complex geometry of the Kowalewski–Painlevé analysis" Invent. Math. , 97 (1989) pp. 3–51 Zbl 0678.58020 |
[a3] | W. Barth, "Abelian surfaces with $(1,2)$-polarization" , Algebraic Geometry, Sendai, 1985 , Advanced Studies in Pure Math. , 10 (1987) pp. 41–84 MR946234 |
[a4] | W. Barth, "Quadratic equations for level-$3$ abelian surfaces" , Abelian Varieties, Proc. Workshop Egloffstein 1993 , de Gruyter (1995) pp. 1–18 MR1336597 |
[a5] | W. Barth, I. Nieto, "Abelian surfaces of type $(1,3)$ and quartic surfaces with $16$ skew lines" J. Algebraic Geom. , 3 (1994) pp. 173–222 MR1257320 Zbl 0809.14027 |
[a6] | Ch. Birkenhake, H. Lange, D. van Straten, "Abelian surfaces of type $(1,4)$" Math. Ann. , 285 (1989) pp. 625–646 MR1027763 Zbl 0714.14028 |
[a7] | Ch. Birkenhake, H. Lange, "Moduli spaces of Abelian surfaces wih isogeny" , Geometry and Analysis, Bombay Colloquium 1992 , Tata Inst. Fundam. Res. (1995) pp. 225–243 MR1351509 |
[a8] | G. Horrocks, D. Mumford, "A rank $2$ vector bundle on $\mathbb{P}^4$ with $15000$ symmetries" Topology , 12 (1973) pp. 63–81 MR382279 Zbl 0255.14017 |
[a9] | K. Hulek, H. Lange, "Examples of abelian surfaces in $\mathbb{P}^4$" J. Reine Angew. Math. , 363 (1985) pp. 200–216 MR0814021 Zbl 0593.14027 |
[a10] | H. Lange, Ch. Birkenhake, "Complex Abelian varieties" , Grundlehren math. Wiss. , 302 , Springer (1992) MR1217487 Zbl 0779.14012 |
[a11] | I. Naruki, "On smooth quartic embeddings of Kummer surfaces" Proc. Japan Acad. , 67 A (1991) pp. 223–224 MR1137912 |
[a12] | V. V. Nikulin, "On Kummer surfaces" Math USSR Izv. , 9 (1975) pp. 261–275 (In Russian) MR429917 Zbl 0325.14015 |
[a13] | S. Ramanan, "Ample divisors on abelian surfaces" Proc. London Math. Soc. , 51 (1985) pp. 231–245 MR0794112 Zbl 0603.14013 |
[a14] | I. Reider, "Vector bundles of rank $2$ and linear systems on algebraic surfaces" Ann. of Math. , 127 (1988) pp. 309–316 MR0932299 Zbl 0663.14010 |
[a15] | P. Vanhaecke, "A special case of the Garnier system, $(1,4)$-polarized Abelian surfaces and their moduli" Compositio Math. , 92 (1994) pp. 157–203 MR1283227 |