Register to reply

Bochner-Weitzenbock formula (-> Laplacian)

Share this thread:
Sajet
#1
Jan2-13, 11:30 AM
P: 48
Hi! I'm trying to understand a proof for the Bochner-Weitzenbock formular. I'm sorry I have to bother you with such a basic question but I've worked at this for more than an hour now, but I just don't get the very first step, i.e.:

[itex]-\frac{1}{2} \Delta |\nabla f|^2 = \frac{1}{2} \sum_{i}X_iX_i \langle \nabla f, \nabla f \rangle[/itex]
Where we are in a complete Riemannian manifold, [itex]f \in C^\infty(M)[/itex] at a point [itex]p \in M[/itex], with a local orthonormal frame [itex]X_1, ..., X_n[/itex] such that [itex]\langle X_i, X_j \rangle = \delta_{ij}, D_{X_i}X_j(p) = 0[/itex], and of course

[itex]\langle \nabla f, X \rangle = X(f) = df(X)[/itex]
[itex]\textrm{Hess }f(X, Y) = \langle D_X(\nabla f), Y \rangle[/itex]
[itex]\Delta f = - \textrm{tr(Hess )}f[/itex]

I've tried to use the Levi-Civita identities, but I'm getting entangled in these formulas and don't get anywhere.

Any help is appreciated.
Phys.Org News Partner Science news on Phys.org
Sapphire talk enlivens guesswork over iPhone 6
Geneticists offer clues to better rice, tomato crops
UConn makes 3-D copies of antique instrument parts
Sajet
#2
Jan3-13, 09:11 AM
P: 48
I got it now :)
dextercioby
#3
Jan3-13, 10:01 AM
Sci Advisor
HW Helper
P: 11,893
You may try to post a solution/sketch of solution for the one interested. That would be nice of you.

Sajet
#4
Feb10-13, 08:32 AM
P: 48
Bochner-Weitzenbock formula (-> Laplacian)

Sorry, i didn't notice the post. In case anyone ever finds this through google or the search function, here it is:

[itex]-\frac{1}{2} \Delta\|\nabla f\|^2 = \frac{1}{2}\text{tr}(\text{Hess}(\langle \nabla f, \nabla f \rangle ))[/itex]
[itex]= \frac{1}{2}\sum_{i=1}^n \langle \nabla_{X_i} \text{grad}\langle \nabla f, \nabla f \rangle, X_i\rangle[/itex] (<- these are the diagonal entries of the representation matrix)
[itex]= \frac{1}{2}\sum_{i=1}^n X_i \langle \text{grad}\langle \nabla f, \nabla f\rangle, X_i\rangle - \langle \text{grad}\langle \nabla f, \nabla f\rangle, \nabla_{X_i} X_i\rangle[/itex] (where the second summand is zero)
[itex]= \frac{1}{2} \sum_{i=1}^n X_i X_i \langle \nabla f, \nabla f\rangle[/itex]


Register to reply

Related Discussions
Laplacian Calculus & Beyond Homework 10
Characteristic functions, Bochner's Thm. Set Theory, Logic, Probability, Statistics 1
Laplacian solution(page 2) to Jackson 1.5 Calculus 1
What is a laplacian? Calculus & Beyond Homework 0
2D Laplacian Calculus 9