Scale-invariance in three-dimensional isotropic turbulence: a paradox and its resolution

Abstract

If the Reynolds number is large enough, turbulence is expected to exhibit scale invariance in an intermediate (‘inertial’) range of wave numbers, as shown by power-law behaviour of the energy spectrum and also by a constant rate of energy transfer through wave number. However, although it has long been known that the first of these is true, there has been little recognition of the fact that, if the second is to hold, then there is a contradiction between the definition of the energy flux (as the integral of the transfer spectrum) and the observed behaviour of the transfer spectrum itself. This is because the transfer spectrum T(k) is invariably found to have a zero crossing at a single point (at k0, say), implying that the corresponding energy flux cannot have an extended plateau but must instead have a maximum value at k = k0. We outline the resulting paradox and note that it may be resolved by the observation that the symmetry of the triadic interactions means that T(k) is not the relevant transfer term in determining the energy flux. Instead the relevant term is a filtered/partitioned version, herein denoted by T+−(k|kc), where k = kc is the cut-off wave number for low/high-pass filtering. It is known from studies of spectral subgrid transfer that T+−(k|kc) is zero over an extended range of wave numbers. As this is the case for quite modest Reynolds numbers, it not only resolves the paradox, but may also shed some light on the ‘embarrassment of success’ of the Kolmogorov theory.