An experimental investigation of forced steady rotating turbulence

This paper presents some of the principal findings of an experimental investigation of forced statistically steady turbulence in a rapidly rotating background. The experiment is conducted in a large cylindrical mixing tank and bulk rotation is produced by two large co-rotating impellers installed near the top and the bottom of the tank. The mean flow motion generated in such a configuration is axisymmetric and close to but not exactly solid-body rotation. The focus of this work is on the region in the vicinity of the rotation axis where the flow is Rayleigh stable. The unique features of this facility allow forcing and thus permit turbulence to be maintained at a certain level without a mean decay. The facility also allows an easy experimental access of statistical quantities. The bulk rotation rates and the amount of turbulence are adjusted to a Rossby number (Ro) of O(1). Under this condition, some physical phenomena commonly seen in solid-body rotating turbulent flows at low Ro are also observed, despite the differences in other parameters. A flow visualisation at one of the testing conditions using pearlescence clearly shows the presence of cyclonic columnar structures. Two-dimensional particle image velocimetry (2DPIV) measurements are carried out in a plane normal to the axis of the rotation and near the middle height of the tank. Symmetry breaking between cyclonic and anti-cyclonic vorticity in the flow is revealed by the skewness of the fluctuating vorticity. It is found that skewness is not a monotonic function of Ro; in this particular experimental arrangement, symmetry is broken to a maximum extent at Ro ∼1.5 based on velocity length scale. Turbulence quantities are also computed along with the turbulent energy dissipation, which is estimated using velocity structure functions and lends support to previous findings that purport that dissipation is suppressed in such flows.