“…͑1͒ suggests, since small-scale drift-wave turbulence ͑indexed as k͒ may drive the largescale fluctuations anisotropically, we model it as a random anisotropic external forcing peaked at high k. Twodimensional turbulence poses yet higher level of complexity than the three-dimensional problem, though, in that additional divergent diagram arises which demands a noise renormalization and, thus, new smallness parameter ␦ϭ2Ϫd as explained by Honkonen and Nalimov. 6 In Sec. II, we formulate the action integral of the forced HME that includes both the nonlocal and the local noises in order to properly treat the renormalizations.…”