Let $\Gamma$ be a compact $C^2$ manifold and suppose that $f_n$ is a non-negative sequence of functions such that:

$(*)\forall t\in (0,T)$ it holds: $\int_{0}^t \int_{\Gamma} f_n \le C_1+C_2t$ for some positive constants $C_1,C_2$.

From $(*)$ we can deduce that ${\vert \vert f_n \vert \vert}_{L^1(0,t,L^1(\Gamma))} \le C_1+C_2T \;\;\forall t\in (0,T)$.

I'm interested in obtaining convergence of $f_n$

Unfortunately, this $L^1(0,t,L^1(\Gamma))$ bound doesn't seem much helpful. If instead I had a bound of $f_n$ in $L^{\infty}(0,t,L^1(\Gamma))$, I could then deduce some convergence using duality arguments. But even in the simpler case in which a sequence is bounded in $L^1(\Gamma)$, then a convergent subsequence may not exist.


  1. Do I miss any theorem in Bochner spaces that provides convergence for sequences bounded in $L^1(0,t,L^1(\Gamma))$?
  2. Is there any other way to take advantage of the inequality in $(*)$ in order to deduce a better bound?

Any help is much appreciated.Thanks in advance!


No, this $L^1$-bound does not give any convergence in $L^1$. Think of the sequence $$ f_n = n^2 \chi_{[0,1/n]^2}, $$ which is uniformly bounded in $L^1(0,1;L^1(0,1))$. It converges in a weak-star sense to the Dirac functional at the origin, and this limit point cannot be represented by an integrable function.

  • $\begingroup$ First of all, thanks a lot for your answer. To be honest, I wasn't expecting $f_n$ to converge in $L^1$ but maybe in some vector-valued measure space (I don't know if I use the right term here and I don't even know if such a space exist). Is there any possibility for $f_n$ to converge utilizing $(*)$? $\endgroup$ – kaithkolesidou Feb 6 at 15:37
  • 1
    $\begingroup$ You would get weak-star convergence in a dual space to a space of continuous function, i.e., in $C([0,T]\times \Gamma)^*$. $\endgroup$ – daw Feb 6 at 20:06

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.