From my lecture notes, on a general Riemannian manifold $(M, g)$, the divergence operator $\operatorname{div}: T^\infty(TM)\rightarrow C^\infty(M)$ is defined as $\operatorname{div}X := \star^{-1}d(x\neg\operatorname{Vol})$ , where $\star$ is the Hodge star operator $C^\infty\rightarrow T^\infty(\bigwedge^nT^\ast M)$ defined as $f\rightarrow f\operatorname{Vol}_g$.

Now I've read this from other place that there is another characterized definition of divergence operator given by:

For $\vec{F}\in C^\infty(U)$ with an open subset $U\subset X$, the function $\operatorname{div}(\vec{F})\in C^\infty$, $$d(\star w(\vec{F})) = \operatorname{div}(\vec{F})dV_X|_U$$

The detail of this definition can be found here.

Now my question. How does the definition given in my lecture imply the characterization property of divergence operator?


If $ \overline{F} = (f_1,f_2,...,f_n) $ then you can identify it locally with the vector field $ X = \sum_i f_i\partial_i $ and thus you have definitions for both of them identical via this. There are several equivalent definitions of divergence, like defining $ div(X)dV = d(i_X dV) $, trace of Levi-Civita connection, etc.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.