11
$\begingroup$

Let $M$ be a Riemann surface (or a higher dimensional manifold) and let's assume that it's geodesically complete. Let $W(t)$ be a Brownian motion on the surface accordingly to the manifold's Laplacian and let $r>0$.

Define the Wiener sausage as:

$$ W_{r}(t):=\{ x\in M: d(x,W(s))\leq r\quad\text{for}\quad 0\leq s\leq t \}. $$

It is known that in $\mathbb{R}^{2}$ and for t sufficiently large and $r$ fixed

$$ \mathbb{E}[\mathrm{vol}(W_{r}(t))]=\frac{2\pi t}{\log(t)}(1+o(1)). $$

Is there any analogue result for a general Riemann surface or at least the hyperbolic space?

Thanks!

--Gabriel

$\endgroup$
5
  • $\begingroup$ Quick scholar googling gave this reference: jstor.org/stable/2244253, which cites a similar result for general two-dimensional Riemann manifold (though the number 2 is missing from the rhs there). $\endgroup$
    – zhoraster
    Mar 22, 2011 at 15:28
  • $\begingroup$ Some further scholar googling gave a similar result: archive.numdam.org/ARCHIVE/CM/CM_1986__60_1/CM_1986__60_1_65_0/… for a dimension $\ge 3$. $\endgroup$
    – zhoraster
    Mar 22, 2011 at 15:29
  • 2
    $\begingroup$ Thanks zhoraster. I'm familiar with these two papers but they focus on the case where $t$ is fixed and $r\to 0$. They essentially proved that in this scenario: $$ \mathbb{E}(\mathrm{vol}(W_{r}(t)))\sim \frac{\pi t}{\log(1/r)}+\frac{\pi t}{2\log(1/r)^2}(1+k-\log(2t)). $$ However, I'm interested in the case where $r$ is fixed and $t\to\infty$ as in the Euclidean case. $\endgroup$
    – ght
    Mar 22, 2011 at 18:32
  • $\begingroup$ BTW, by just comparing with the Euclidean case you see that the behavior is quiet different in these two cases. $\endgroup$
    – ght
    Mar 22, 2011 at 18:34
  • $\begingroup$ Welcome, Gabriel! $\endgroup$
    – Jon Bannon
    Mar 22, 2011 at 19:26

1 Answer 1

3
$\begingroup$

I just found out that the case $r$ fixed and $t\to\infty$ for simply connected symmetric manifolds of non-positive sectional curvature and dimension $d\geq 3$, and strictly negative curvature for dimension $d=2$, was solved by Chavel and Feldman in "The Wiener Sausages and a Theorem of Spitzer in Riemannian Manifolds", Probability and Harmonic Analysis, New York, pp. 45-60, 1986.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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