# User:Maximilian Janisch/latexlist/Algebraic Groups/Kähler-Einstein metric

A Kähler metric on a complex manifold (or orbifold) whose Ricci tensor $\operatorname { Ric } ( \omega )$ is proportional to the metric tensor:

\begin{equation} \operatorname { Ric } ( \omega ) = \lambda \omega \end{equation}

This proportionality is an analogue of the Einstein field equation in general relativity. The following conjecture is due to E. Calabi: Let $N$ be a compact connected complex manifold and $c _ { 1 } ( M ) _ { R }$ its first Chern class; then

a) if $c _ { 1 } ( M ) _ { R } < 0$, then $N$ carries a unique (Ricci-negative) Kähler–Einstein metric $4$ such that $\operatorname { Ric } ( \omega ) = - \omega$;

b) if $c _ { 1 } ( M ) _ { R } = 0$, then any Kähler class of $N$ admits a unique (Ricci-flat) Kähler–Einstein metric such that $\operatorname { Ric } ( \omega ) = 0$.

This conjecture was solved affirmatively by T. Aubin [a1] and S.T. Yau [a8] via studies of complex Monge–Ampère equations, and Kähler–Einstein metrics play a very important role not only in differential geometry but also in algebraic geometry. The affirmative solution of this conjecture gives, for instance, the Bogomolov decomposition for compact Kähler manifolds with $c _ { 1 } ( M ) _ { R } = 0$. It also implies (see [a2], [a3]):

1) Any Kähler manifold homeomorphic to $C ^ { \prime } D ^ { \prime }$ is biholomorphic to $C ^ { \prime } D ^ { \prime }$. Any compact complex surface homotopically equivalent to $C P ^ { 2 }$ is biholomorphic to $C P ^ { 2 }$.

2) In the Miyaoka–Yau inequality $c _ { 1 } ( S ) ^ { 2 } \leq 3 _ { C 2 } ( S )$, for a compact complex surface $5$ of general type, equality holds if and only if $5$ is covered by a ball in $c ^ { 2 }$.

For a Fano manifold $N$ (i.e., $N$ is a compact complex manifold with $c _ { 1 } ( M ) _ { R } > 0$), let $k$ be the identity component of the group of all holomorphic automorphisms of $N$. Let $\varepsilon$ be the set of all Kähler–Einstein metrics $4$ on $N$ such that $\operatorname { Ric } ( \omega ) = \omega$. If $E \neq \emptyset$, then $\varepsilon$ consists of a single $k$-orbit (see [a5]). Moreover, the following obstructions to the existence of Kähler–Einstein metrics are known (cf. [a5], [a6]):

Matsushima's obstruction. If $E \neq \emptyset$, then $k$ is a reductive algebraic group (cf. also Reductive group).

Futaki's obstruction. If $E \neq \emptyset$, then Futaki's character $F _ { M } : G \rightarrow C ^ { * }$ is trivial.

Recently (1997), G. Tian [a7] showed some relationship between the existence of Kähler–Einstein metrics on $N$ and stability of the manifold $N$, and gave an example of an $N$ with no non-zero holomorphic vector fields satisfying $E = \emptyset$.

The Poincaré metric on the unit open disc $\{ z \in C : | z | < 1 \}$ (cf. Poincaré model) and the Fubini–Study metric on $C ^ { \prime } D ^ { \prime }$ are both typical examples of Kähler–Einstein metrics. For more examples, see Kähler–Einstein manifold.

For the relationship between Kähler–Einstein metrics and multiplier ideal sheaves, see [a4]. See, for instance, [a2] for moduli spaces of Kähler–Einstein metrics. Finally, Kähler metrics of constant scalar curvature and extremal Kähler metrics are nice generalized concepts of Kähler–Einstein metrics (cf. [a2]).

How to Cite This Entry:
Maximilian Janisch/latexlist/Algebraic Groups/Kähler-Einstein metric. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Maximilian_Janisch/latexlist/Algebraic_Groups/K%C3%A4hler-Einstein_metric&oldid=44019