Formula for sectional curvature

54 Views Asked by At

I am so confused about the following problem (maybe I have made some silly conceptual mistakes) The reference I used is Lee's introduction to Riemannian manifold. enter image description here

There are two theorems in this book.

enter image description here enter image description here

According to my understanding, the sectional curvature is a scalar, which can be computed by any basis of the tangent space by formula in theorem 8.29 (i.e. does not depend on the basis). Now, in exercise 8.30, $\lambda$ is a constant, substitute it into theorem 7.30 shows $f$ is a constant.

Then it seems (7.44) and (7.46) give different answers. 7.44 and 8.29 implies $\tilde{sec}(\Pi)=\lambda sec(\pi)$ while (7.46) implies $\tilde{sec}(\Pi)=\lambda^{-1} sec(\pi)$. How can this happen?

Does anyone have any ideas or comments?

1

There are 1 best solutions below

6
On BEST ANSWER

Applying $(7.44)$ with constant $f$ shows that $\widetilde{\text{Rm}}=\lambda\cdot\text{Rm}$. So, by using formula $(8.28)$, we get that for a plane $\Pi$ spanned by two linearly independent vectors $v,w$, \begin{align} \widetilde{\text{sec}}(\Pi)&=\frac{\widetilde{\text{Rm}}(v,w,w,v)}{|v\wedge w|_{\widetilde{g}}^2}=\frac{\lambda\cdot \text{Rm}(v,w,w,v)}{\lambda^2\cdot|v\wedge w|_g^2}=\frac{1}{\lambda}\cdot\text{sec}(\Pi), \end{align} which is exactly what is claimed. I think you forgot to account for the different norm in the denominator.