Comparison between temporal and spatial transverse velocity increments in a turbulent plane jet

Longitudinal and transverse velocity increments are measured on the centreline of a turbulent plane jet at a Taylor microscale Reynolds number R_λ of about 1000. The transverse increments are obtained both spatially, using two parallel hot wires separated in the spanwise (shearless) direction, and temporally, using an X-wire probe. In the latter case, Taylor's hypothesis is used to convert temporal increments of the transverse velocity fluctuation to spatial ones. For both cases, local isotropy is approximately satisfied in the dissipative range. Local isotropy is also checked in the inertial range, using the well-known relationship between the second-order moments of the longitudinal and transverse velocity increments. The second-order moments of the spatial transverse increments satisfy this relationship more closely than those of the temporal transverse increments. It is conjectured that, for globally anisotropic turbulent shear flows at moderate R_λ, the perceived inertial range scaling of transverse increments is very dependent on the method used to measure these increments. The current observations tend to suggest that the spatial transverse increments, measured in a spanwise direction, are less anomalous with respect to (1941) scaling than the temporal increments formed with the transverse velocity components.