2
$\begingroup$

Consider the Laplace-Beltrami operator $\Delta_g$ on compact Riemannian manifold $(M,g)$, then $e^{it\sqrt{-\Delta_g}}f$ is the solution of the following Cauchy problem. $$ i\partial_tu=\sqrt{-\Delta_g}u, ~u|_{t=0}=f. $$ When $t$ is small, we have $$ e^{it\sqrt{-\Delta_g}}=Q(t)+R(t) $$ where the kernel of $Q(t)$ can be expressed by an oscillatory integral, and the reminder term has smooth kernel. My question is about what happens for large $t$, I'm not expecting good results for general manifold, however, if we just assume that $M$ is the sphere $S^{n}$ with the standard metric, can we say more about $e^{it\sqrt{-\Delta_g}}$ for large $t$? Especially, I want to know if there are explicit expressions of its kernel at any time $t>0$. Since the geodesic flow on the sphere is periodic with a common minimal period $2\pi$, is there also some periodic property of the group $e^{it\sqrt{-\Delta_g}}$?

Thanks very much for any comment or reference.

$\endgroup$
4
  • $\begingroup$ Only in the high frequency limit do the geodesic flow approximate solutions to the wave equation (semiclassical analysis). This is evident in the fact that the spectrum of the Laplacian on the sphere are not all "nice numbers" except in 1D (the eigenvalues of $\triangle_g$ on $\mathbb{S}^N$ are $\ell(\ell + N - 1)$ where $\ell \in \mathbb{N}\cup \{0\}$, so in general their square-roots do not have rational ratios). $\endgroup$ Commented Jan 21, 2014 at 8:46
  • $\begingroup$ On the other hand, if you relax the notion of periodicity to allow some errors, your intuition is in fact true. The eigenvalues of the Laplacian on the sphere cluster around perfect squares, so in that sense the periodicity plays some role. $\endgroup$ Commented Jan 21, 2014 at 12:10
  • $\begingroup$ Also, starting from the link in my previous comment, I came across this paper which you might find interesting/useful. $\endgroup$ Commented Jan 21, 2014 at 12:18
  • $\begingroup$ @Willie Wong,thanks for the reference, I also found that in chapter 29 of Hormander's book, The Analysis of Linear Partial Differential Operators, he discussed the case of the periodic Halmilton flow, and showed the singularity of the operator at any time, though I had to take more time to get understand of the details. $\endgroup$
    – Tomas
    Commented Jan 21, 2014 at 22:35

0

You must log in to answer this question.

Browse other questions tagged .