ScalingStacks

3.2 Monge-Ampère manifolds and duality of torus fibrations [03QN]

Original official author HTML, exact retained edition. Historical TeX conversion verdicts remain unchanged. Cited-edition alignment and mathematical self-containment are not assessed.

Complete original source context · Original author HTML

3.2 Monge-Ampère manifolds and duality of torus fibrations

In this section we propose a mathematical language for the geometric mirror symmetry, understood as a duality of torus fibrations.

Definition 2

A Monge-Ampère manifold is a triple (Y,g,∇)(Y,g,\nabla), where (Y,g)(Y,g) is a smooth Riemannian manifold with the metric gg, and ∇\nabla is a flat connection on TYT_{Y} such that:

a) ∇\nabla defines an affine structure on YY.

b) Locally in affine coordinates (x1,…,xn)(x_{1},...,x_{n}) the matrix ((gi​j))((g_{ij})) of gg is given by ((gi​j))=((∂2K/∂xi​∂xj))((g_{ij}))=((\partial^{2}K/\partial x_{i}\partial x_{j})) for some smooth real-valued function KK.

c) The Monge-Ampère equation d​e​t​((∂2K/∂xi​∂xj))=c​o​n​s​tdet((\partial^{2}K/\partial x_{i}\partial x_{j}))=const is satisfied.

Monge-Ampère manifolds were studied (under a different name) in [CY] where is was proven that if YY is compact then its finite cover is a torus.

Let us consider a (non-compact) example motivated by the mirror symmetry for K3 surfaces (see also [GW]). Let SS be a complex surface endowed with a holomorphic non-vanishing volume form v​o​lSvol_{S}, and π:S→C\pi:S\to C be a holomorphic fibration over a complex curve CC, such that fibers of π\pi are non-singular elliptic curves.

We define a metric gCg_{C} on CC as the Kähler metric associated with the (1,1)(1,1)-form π∗​(v​o​lS∧v​o​l¯S)\pi_{\ast}(vol_{S}\wedge\overline{vol}_{S}). Let us choose (locally on CC) a basis (γ1,γ2)(\gamma_{1},\gamma_{2}) in H1​(π−1​(x),𝐙),x∈CH_{1}(\pi^{-1}(x),{{\bf Z}}),x\in C. We define two closed 1-forms on CC by the formulas

αi=Re(∫γivolS),i=1,2.\alpha_{i}=Re(\int_{\gamma_{i}}vol_{S}),i=1,2.

It follows that αi=d​xi\alpha_{i}=dx_{i} for some functions xi,i=1,2x_{i},i=1,2. We define an affine structure on CC, and the corresponding connection ∇\nabla, by saying that (x1,x2)(x_{1},x_{2}) are affine coordinates. One can check directly that (C,gC,∇)(C,g_{C},\nabla) is a Monge-Ampère manifold. In a typical example of elliptic fibration of a K3 surface, one gets C=𝐂𝐏1∖{z1,…,z24}C={{\bf C}}{\bf P}^{1}\setminus\{z_{1},...,z_{24}\}, where {z1,…,z24}\{z_{1},...,z_{24}\} is a set of distinct 2424 points in 𝐂𝐏1{{\bf C}}{\bf P}^{1}.

Returning to the general case, we can restate a portion of our conjectures by saying that the smooth part of the Gromov-Hausdorff limit of a maximally degenerate family of Calabi-Yau manifolds is a Monge-Ampère manifold with an integral affine structure.

There is a well-known duality on local solutions of the Monge-Ampère equation.

Lemma 2

Let U⊂𝐑nU\subset{{\bf R}}^{n} be a convex open domain in 𝐑n{{\bf R}}^{n} equipped with the standard affine coordinates (x1,…,xn)(x_{1},...,x_{n}), and K:U→𝐑K:U\to{{\bf R}} be a convex function satisfying the Monge-Ampère equation. Then the Legendre transform K^​(y1,…,yn)=m​a​xx∈U​(∑ixi​yi−K⁡(x1,…,xn))\widehat{K}(y_{1},...,y_{n})=max_{x\in U}(\sum_{i}x_{i}y_{i}-K(x_{1},...,x_{n})) also satisfies the Monge-Ampère equation.

Proof. The graph of L=d​KL=dK is a Lagrangian submanifold in T∗​𝐑n=𝐑n⊕(𝐑n)∗T^{\ast}{{\bf R}}^{n}={{\bf R}}^{n}\oplus({{\bf R}}^{n})^{\ast}. Let p1p_{1} and p2p_{2} be the natural projections to the direct summands. They are local diffeomorphisms. Since KK is defined up to the adding of an affine function, the graph itself is defined up to translations. The Monge-Ampère equation corresponds to the condition p1∗​(v​o​l𝐑n)=p2∗​(v​o​l𝐑∗n)p_{1}^{\ast}(vol_{{\bf R}^{n}})=p_{2}^{\ast}(vol_{{\bf R}^{\ast n}}), where v​o​l𝐑nvol_{{\bf R}^{n}} (resp. v​o​l𝐑∗nvol_{{\bf R}^{\ast n}}) denotes the standard volume form on 𝐑n{\bf R}^{n} (resp. 𝐑∗n{\bf R}^{\ast n}). The manifold LL can be considered as a graph of d​K^d\widehat{K}. Thus K^\widehat{K} satisfies the Monge-Ampère equation as well. The Lemma is proved. ■\blacksquare

The manifold LL carries a Riemannian metric gLg_{L} induced by the indefinite metric ∑id​xi​d​yi\sum_{i}dx_{i}\,dy_{i} on 𝐑n⊕(𝐑n)∗{{\bf R}}^{n}\oplus({{\bf R}}^{n})^{\ast}. This metric is given by the matrix (∂2K/∂xi​∂xj)(\partial^{2}K/\partial x_{i}\partial x_{j}) in coordinates (x1,…,xn)(x_{1},...,x_{n}), and by the matrix (∂2K^/∂yi​∂yj)(\partial^{2}\widehat{K}/\partial y_{i}\partial y_{j}) in the dual coordinates. Thus on LL we have a metric, and two affine structures (pullbacks of the standard affine structures on the coordinate spaces). Hence we have two structures of the Monge-Ampère manifold on LL. It is easy to see that the local pictures can be glued together. This leads to the following result.

Proposition 1

For a given Monge-Ampère manifold (Y,gY,∇Y)(Y,g_{Y},\nabla_{Y}) there is a canonically defined dual Monge-Ampère manifold (Y∨,gY∨,∇Y∨)(Y^{\vee},g_{Y}^{\vee},\nabla_{Y}^{\vee}) such that (Y,gY)(Y,g_{Y}) is identified with (Y∨,gY∨)(Y^{\vee},g_{Y}^{\vee}) as Riemannian manifolds, and the local system (TY∨,∇Y∨)(T_{Y^{\vee}},\nabla_{Y}^{\vee}) is naturally isomorphic to the local system dual to (TY,∇Y)(T_{Y},\nabla_{Y}).

Corollary 1

If ∇Y\nabla_{Y} defines an integral affine structure on YY (i.e. the holonomy of ∇Y\nabla_{Y} belongs to G​L​(n,𝐙)GL(n,{{\bf Z}})), then ∇Y∨\nabla_{Y}^{\vee} defines an integral affine structure on Y∨Y^{\vee}. As the dual covariant lattice one takes the lattice (TY𝐙)∨(T^{{\bf Z}}_{Y})^{\vee}, which is dual to TY𝐙T_{Y}^{{\bf Z}} with respect to the metric gYg_{Y}.

Now we can state the geometric counterpart of the mirror symmetry conjecture.

Conjecture 3

Smooth parts of maximal degenerations of dual families of Calabi-Yau manifolds are dual Monge-Ampère manifolds with dual integral affine structures.

Monge-Ampère manifolds with integral affine structures are real analogs of Calabi-Yau manifolds. In fact the mirror duality in the sense of this section holds for a larger class of manifolds. We define an AK-manifold (AK stands for affine and Kähler) as in the Definition 2, but dropping the condition c) (Monge-Ampère equation), see also [CY]. The reader can check easily that all constructions of this section, including the duality of torus fibrations hold for AK-manifolds as well.

Remark 6

The idea to use the Legendre transform for the purposes of mirror symmetry was around for some time (see for example [H], [Le]).

Remark 7

In our description of geometric mirror symmetry we ignore the B-fields. In what follows we will always assume that B=0B=0.

Original mathematics by the credited authors. Source-backed reader collection; mathematical self-containment is not assessed.