ScalingStacks

5.1. The 4 –manifold M ϵ [02HH]

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

5.1. The 44–manifold MϵM_{\epsilon}

Let 𝝎¯ϵgh\bm{\underline{\omega}}^{\textup{gh}}_{\epsilon} be the hyperkähler triple defined in (4.6). By Lemma 4.9 for ϵ>0\epsilon>0 small enough we think of 𝝎¯ϵgh\bm{\underline{\omega}}^{\textup{gh}}_{\epsilon} as defined on MϵghM^{\textup{gh}}_{\epsilon}, a smooth manifold with boundary obtained by restricting the line bundle PP to the complement of (arbitrarily) small balls centred at the punctures and then taking the quotient by the involution τ~\tilde{\tau}. The boundary of MϵghM^{\textup{gh}}_{\epsilon} has n+8n+8 components, each of which has a collar neighbourhood diffeomorphic either to H2​mj−4/ℤ2H^{2m_{j}-4}/\mathbb{Z}_{2} for j=1,…,8j=1,\dots,8, or HkiH^{k_{i}} for i=1,…,ni=1,\dots,n.

For each j=1,…,8j=1,\dots,8 let MjM_{j} be the smooth 44–manifold underlying a DmjD_{m_{j}} ALF space and for each i=1,…,ni=1,\dots,n let NiN_{i} be the smooth manifold underlying an Aki−2A_{k_{i}-2} ALF space. We construct a smooth 44–manifold MϵM_{\epsilon} by cutting the ends of MjM_{j} and NiN_{i} and gluing the resulting manifolds with boundary to MϵghM^{\textup{gh}}_{\epsilon} in a neighbourhood of qjq_{j} or ±pi\pm p_{i}, respectively.

We will construct an approximate hyperkähler structure on MϵM_{\epsilon} in the next subsection. Here we pause for a moment to determine the Betti numbers of MϵM_{\epsilon}. While we will not use this result in an essential way in the rest of the paper, it is interesting to note how the balancing condition (4.1) appears naturally in the calculation of the Euler characteristic of MϵM_{\epsilon}.

Proposition 5.1.

The Betti numbers of the compact orientable 44–manifold MϵM_{\epsilon} are

b1​(Mϵ)=0,b2+​(Mϵ)=3,b2−​(Mϵ)=22.b_{1}(M_{\epsilon})=0,\qquad b_{2}^{+}(M_{\epsilon})=3,\qquad b_{2}^{-}(M_{\epsilon})=22.
Proof.

Decompose MϵM_{\epsilon} into the union of a piece P/τ~P/\tilde{\tau}, an Aki−1A_{k_{i}-1} ALF space for each i=1,…,ni=1,\dots,n and a DmjD_{m_{j}} ALF space for each j=1,…,8j=1,\dots,8. These pieces are identified along their common boundaries, which are homology spheres. Since all components have vanishing first Betti number, the reduced Mayer–Vietoris sequence yields b1​(Mϵ)=0b_{1}(M_{\epsilon})=0. The Euler characteristic is also easily calculated:

χ⁡(Mϵ)=χ⁡(P/ℤ2)+∑i=1nχ⁡(Aki−1)+∑j=18χ⁡(Dmj)=0+∑i=1nki+∑j=18mj+8=24\chi(M_{\epsilon})=\chi(P/\mathbb{Z}_{2})+\sum_{i=1}^{n}{\chi(A_{k_{i}-1})}+\sum_{j=1}^{8}{\chi(D_{m_{j}})}=0+\sum_{i=1}^{n}{k_{i}}+\sum_{j=1}^{8}{m_{j}}+8=24

by the balancing condition (4.1).

It remains to calculate the signature τ⁡(Mϵ)\tau(M_{\epsilon}). Below we will construct a definite triple 𝝎¯ϵ\bm{\underline{\omega}}_{\epsilon} on MϵM_{\epsilon} which is close to define a hyperkähler structure. By changing basis of Λ+​T∗​Mϵ\Lambda^{+}T^{\ast}M_{\epsilon} one can always deform this triple to a genuine S​U​(2)SU(2)–structure (without requiring any differential constraint). In particular, MϵM_{\epsilon} can be endowed with an almost complex structure JJ with c1​(Mϵ,J)=0c_{1}(M_{\epsilon},J)=0. Since c2​(Mϵ,J)=χ⁡(Mϵ)=24c_{2}(M_{\epsilon},J)=\chi(M_{\epsilon})=24, Hirzebruch’s Signature Theorem and the equality of characteristic classes p1=c12−2​c2p_{1}=c_{1}^{2}-2c_{2} yield τ⁡(Mϵ)=−16\tau(M_{\epsilon})=-16. ∎

Remark.

In Remark 4.8 we noted that the case where n=0n=0 and mj=2m_{j}=2 for all j=1,…,8j=1,\dots,8 reduces to the usual Kummer construction. Hence we know that MϵM_{\epsilon} is diffeomorphic to the K3 surface in this special case. It seems likely one can prove that the diffeomorphism type of MϵM_{\epsilon} does not depend on the configuration of punctures satisfying the balancing condition (4.1). Since we are going to construct a hyperkähler metric on MϵM_{\epsilon}, the calculation of the Betti numbers will anyway imply that MϵM_{\epsilon} is always diffeomorphic to the K3 surface.

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