Let $\Omega$ be a bounded (at least) Lipschitz domain in $\mathbb{R}^{N}$. Its boundary $\partial\Omega$, if I'm right, is $(N-1)$-dimensional object embedded in $\mathbb{R}^N$.

We can define Sobolev spaces $H^1(\partial\Omega)$. My question is what dimension should I think of for $\partial\Omega$? I want to use Sobolev inequalities on the boundary (= manifolds) where I need to specify the dimension obviously. So should I say that Sobolev spaces over $\partial\Omega$ are spaces on a $(N-1)$-dimensional space or an $N$-dimensional space? The former seems right, but I don't know how to justify this. Thanks.

Edit: my main issue is that to define $H^1(\partial\Omega)$, you need to define the tangential gradient, which is defined as $\nabla_T f = \nabla \tilde f - \nabla \tilde f\cdot \nu$ where $\nu$ is the unit normal and $\tilde f$ is an extension of $f$ to an open set surrounding $\partial\Omega$. This tangential gradient has $N$ components, not $N-1$. SO how can $H^1(\partial\Omega)$ be on a $N-1$-dimensional set?


The number of components is not the dimension. Here, because there is 1 "constraint" (that the normal component of the tangential derivative is zero), the boundary is indeed a codimension 1 submanifold.

You probably don't actually believe the number of components is the dimension, but this particular scenario has confused you. For example, in $\mathbb{R}^n$ you could have the constraint $x_1 + x_2 = 0$. The set of all points satisfying the constraint is a hyperplane of course.


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.