Extremal metrics and lower bound of the modified K-energy (Q747610): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Set profile property. |
||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank |
Revision as of 01:06, 5 March 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Extremal metrics and lower bound of the modified K-energy |
scientific article |
Statements
Extremal metrics and lower bound of the modified K-energy (English)
0 references
16 October 2015
0 references
Extremal Kähler metrics in a Kähler class \([\omega]\) are metrics, such that the \((1,0)\)-gradient of their scalar curvature is a holomorphic vector field. In particular, constant scalar curvature Kähler (cscK) metrics are extremal. The cscK metrics are critical points of the Mabuchi K-energy. Chi Li proved that if \([\omega]=c_1(L)\) for some line bundle \(L\), then indeed cscK metrics achieve the absolute minimum of the K-energy. Extremal metrics are known to be critical points of a modified K-energy functional. Therefore it is natural to ask whether a Li-like theorem holds in this setting. In this paper, by adapting Li's technique, the authors prove that this is indeed the case (Theorem A). The strategy (a modification of a strategy originally due to Donaldson in the case of cscK metrics with discrete automorphism group of \((X,L)\)) is as follows: One takes the extremal metric, considers Kodaira embeddings of \(X\) using \(L^k\) into projective space, takes the pullbacks of the resulting Fubini-Study metrics \(\omega_k\), shows that they are ``almost'' \(\sigma\)-balanced, notices that \(\sigma\)-balanced metrics are minimizers of functionals \(Z_\sigma\) that converge to the modified K-energy as \(k\rightarrow \infty\) and completes the proof using delicate estimates involving the Bergman kernel.
0 references
extremal metrics
0 references
K-energy
0 references
discrete automorphism group
0 references
Kodaira embeddings
0 references
\(\sigma\)-balanced
0 references
Bergman kernel
0 references