“…Summing over all indices (i, j, k) ∈ Λ 0 j and applying Hölder's inequality with exponents p/τ > 1 and p/(p − τ ) one finds Obviously, the sums on the right hand side converge if, and only if, α ∈ 0, γ ∧ ν d θ := 2θ ′ (p − 1)/(p − 2) − dp/(p − 2), where θ/p + θ ′ /p ′ = d with 1/p + 1/p ′ = 1. Moreover, by Lemma 2.1(i) we can choose a sequenceφ n ⊆ C ∞ 0 (G (1) ) approximating ϕ • φ in H 1 p,θ−p (G (1) ). Furthermore, a consequence of our assumptions is, that for all n ∈ N, with probability one, the equality u(t, ·),φ n = u(0, ·),φ n + t 0 f (s, ·),φ n ds + Hence, since H 1 p,θ−p (G (1) ) is continuously embedded in H 1 p ′ ,θ ′ −p ′ (G (1) ) ∩ L p ′ ,θ ′ (G (1) ) ∩ Lp ,θ (G (1) ), the assertion follows for the case p > 2.…”