What is a reference for the following result (which appears to be well-known in measure theory)?
Any non negative Radon measure can be decomposed uniquely into the sum of an absolutely continuous measure and a singular term with respect to the harmonic capacity. The absolute continuous term itself can be decomposed (not uniquely) to a function in $L^1$ and a function in $H^{-1}$ (dual of $H_0^1$).