I am trying to understand section about dual of sobolev spaces from Adam's book "Sobolev Spaces". As I understand it he states that every element in the dual of $W^{k,p}(\Omega)$, $\Omega \subset \mathbb{R}^{n}$ open, can be represented by a vector $v=(v_1,v_2,....,v_N)$ where each $v_i$ is in $L^{q}$ and q is the conjugate exponent of p.N is the number of multi derivate indices that depend on k and n. This also represents an element in the space of distributions $(D(\Omega))^{'}$. But however he says that this representation need not be unique. Can you give an example where there are more than one such representation for elements in the dual $(W^{k,p}(\Omega))^{'}$?
Edited:
Is it possible that there are two different elements in $(W^{k,p}(\Omega))^{'}$, $T_1$ and $T_2$ such that they agree on $C_0^{\infty}(\Omega)$. ie is it possible that $T_1{\phi}=T_2{\phi}$ for all $\phi$ in $C_0^{\infty}(\Omega)$ but $T_1{u} \neq T_2{u}$ for some $u$ in $W^{k,p}(\Omega)$.
The relevant section can be found in chapter 3 of the book Adams, R.A. and Fournier, J.J.F, Sobolev Spaces, 2nd ed., Academic Press, 2003.
The non-uniqueness refers to the vector $v$. Note that for $f\in W^{1,p}(\Omega)$ and $\phi \in C_0^1(\Omega)$ we have $$\int_\Omega \nabla f \cdot \phi d x = - \int_\Omega f \mathop{div} \phi d x.$$ In other words, if you find a function $\phi \in C_0^1(\Omega)$ with $\mathop{div} \phi = 0$, (and there are many of those) you can add it to the respective components of $v$ without changing the operator. The same of course also can be done in the higher derivatives.