1
$\begingroup$

Let $(M,g)$ be a Riemannian manifold and $V$ a unit Killing vector field on it. Under what condition on curvature tensor the following equation hold: $$\nabla_VQ=0,$$ where $Q$ is the Ricci operator defined as $g(QX,Y)=\rho(X,Y)$.

Update1: Einstein metrics that admit a unit Killing vector field satisfies the above equation.

Update2: Can anybody give an example other than Einstein manifolds that satisfy the above equation?

Improve this question
$\endgroup$
1
  • $\begingroup$ Cartesian product of a sphere with a hyperbolic manifold. $\endgroup$
    – Deane Yang
    Commented Aug 3, 2017 at 13:35

1 Answer 1

Reset to default
2
$\begingroup$

If $V$ is a Killing vector, then necessarily its Lie derivative annihilates all curvature tensors, including the Ricci operator, namely $$ \mathcal{L}_V Q = \nabla_V Q + Q J_V - J_V Q = 0 , $$ where $J_V$ is the ("Jacobian") endomorphism on tangent vectors defined by $\nabla_U V = J_V U$. Thus, the condition that you want seems to be just $Q J_V - J_V Q = 0$. I don't see how the unit condition on $V$ enters.

Improve this answer
$\endgroup$
4
  • $\begingroup$ Can be obtained this condition on Riemann curvature tensor? $\endgroup$
    – C.F.G
    Commented Aug 2, 2017 at 17:22
  • $\begingroup$ Yes. Another way to think about this is that a Killing vector field is just an infinitesimal variation of a family of isometries. But the pull back of the metric under the isometry is just the metric itself, so the pullback of any local invariant, such as the curvature tensor and its covariant derivatives will be equal to the original value. Therefore, the infinitesimal variation has to be zero. $\endgroup$
    – Deane Yang
    Commented Aug 2, 2017 at 20:15
  • $\begingroup$ @C.F.G I'm not sure I understand your question. A condition on the Ricci operator is by definition also a condition on the Riemann tensor. Maybe you are looking for a condition independent of $V$? $\endgroup$
    – Igor Khavkine
    Commented Aug 3, 2017 at 4:12
  • $\begingroup$ You are right. I want to know if we assume that $(M,g)$ is Einstein manifold then can be determine curvature tensor completely? Answer of my second update is positive or not? $\endgroup$
    – C.F.G
    Commented Aug 3, 2017 at 6:05

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .