Sign up ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Can anyone help me and prove that in the case of the Ricci flow, the symmetries of the flow are scalings and diffeomorphisms?

Thanks for your time.

share|cite|improve this question
crosspost:… –  Sepideh Bakhoda May 26 '13 at 14:56
any textbook will have this proof, see for example… –  Carlo Beenakker May 26 '13 at 15:31
@Carlo: Most textbooks on Ricci flow will indeed show that scalings and diffeomorphisms are symmetries of the PDE, but I am not so sure that there are many books that show the converse, namely, that any symmetry of the PDE system (once this has been properly defined) is necessarily a combination of scalings and diffeomorphisms. –  Robert Bryant May 26 '13 at 15:53
Could someone say what the definition of a symmetry of the flow is? My reaction to the question was: isn't that the definition of a symmetry? –  Deane Yang May 26 '13 at 18:22
@Deane: I took 'the Ricci flow' to mean the PDE system itself which one can think of as a submanifold of an appropriate jet bundle $J$ over $M\times\mathbb{R}$. Then a 'symmetry' would a self-diffeomorphism of $J$ that carries solutions to solutions (thought of via their $k$-jet graphs). @Carlo: You probably want to be more precise about what 'operations' you allow. Let $\mathcal{S}$ be the set of all Ricci-flat metrics on $\mathbb{R}^n$ and let $\sigma:\mathcal{S}\to\mathcal{S}$ be any mapping whatsoever. This is an 'operation that transforms one solution to another solution', no? –  Robert Bryant May 27 '13 at 0:43

1 Answer 1

I apologise first for submitting this comment as an answer (not enough MO-cred to comment) and second for submitting a question rather than a comment:

What I can gather from the comments above is that the symmetries of the Ricci flow are completely understood: Every `symmetry' of the Ricci flow is a spatial diffeomorphism combined with scaling (and every such combination is a symmetry). My question is whether the problem is also solved for the mean curvature flow (of, say, hypersurfaces of Euclidean space)?

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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