Re: Laplaces equation and measures
- From: craig <ctcowan@xxxxxxxxxxx>
- Date: Wed, 29 Aug 2007 15:15:17 EDT
craig <ctcowan@xxxxxxxxxxx> writes:
Suppose Omega is some open bounded set in R^n (nOmega.
=3)
Let - \Delta E = \mu in Omega
where \mu is some measure with compact support in
on the support of \mu.
I am interested in when I can say that E is finite
the unit ball (which we
Example 1
Suppose \mu is the surface measure asscociated to
assume is contained in Omega). Then a directcalculation shows that E is
finite on the surface of the unit ball.on the support.
Example 2
If \mu is a Dirac mass then we know E is not finite
dimeansional manifold
So I would assume if \mu is supported on some n-1
then E is finite on the support.
What about the sum (surface measure on a ball) +
(Dirac mass at a point on the
surface)? The support is still the surface of the
ball, but E is no longer
finite on the support.
I guess by above i meant to add a condition so the measure behaved like a surface measure; but as your example below shows it doesn't matter.
Without looking at examples one can even see what you point out below. Say n big enough and p small enough such that
there is a u in W^{2,p} \ L^\infty and then take
f:= - \Delta u \in L^p.
Thanks for the counter example.
NEW QUESTION:
Let A \subset \subset Omega.
Let \mu be the Hausdorf measure associated with A ie. if dim(A)=a then \mu(E):=H^{a}(A \cap E) where I am assuming that H^{a} (A)<infty.
I still think that if a \ge n-1 then we have desired result but ?
If the above true then is it true for some a < n-1?
.
You could also find a finite measure that is
absolutely continuous wrt
Lebesgue measure on R^n but where E is not finite on
the support.
Consider the spherically symmetric case, and take the
measure mu with
density r^(p-n-2) where n > p > 2. Note that this is
integrable.
If I'm not mistaken, you get a solution E(r) =
r^(p-n)/((p-2)(n-p))
with goes to infinity as r -> 0.
--
Robert Israel
israel@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Department of Mathematics
http://www.math.ubc.ca/~israel
University of British Columbia Vancouver,
BC, Canada
- References:
- Re: Laplaces equation and measures
- From: Robert Israel
- Re: Laplaces equation and measures
- Prev by Date: Re: diophantine equation 1/x + 1/y = 1/n
- Next by Date: Re: Finding shortest path into unweighted undirected graph
- Previous by thread: Re: Laplaces equation and measures
- Next by thread: Re: Laplaces equation and measures
- Index(es):
Relevant Pages
|