Erratum: The structure and budget of turbulent kinetic energy in front of a wall-mounted cylinder (Journal of Fluid Mechanics. (2017) 827 ( 285-321) DOI: 10.1017/jfm.2017.486)

In Schanderl et al. (2017), incorrect particle image velocimetry (PIV) results were presented owing to mistaken use of the wrong calibration file in the processing of the data. None of the conclusions are affected by this unfortunate error, but the quantitative PIV results need updating with table 2 and figures 7, 8, 11, 12, 15 and 18 replaced with the following. The spatial resolution was 0:00295D (instead of 0:0038D as given in the original article) which corresponds to 339 data points per cylinder diameter, which affected both the spatial position and the amplitudes of the measured flow quantities. While in the results published in the original article the measured position of the horseshoe vortex was by 0:03D more upstream, it turned out that with the correct calibration, it is by 0:03D more downstream of the simulated horseshoe vortex centre. The positions of the other critical points in the symmetry plane, as given in table 2, have changed as well. The magnitudes of the velocity and its statistics are reduced. For example, it was reported that the measured peak value of the in-plane turbulent kinetic energy was 30% larger than the in-plane turbulent kinetic energy simulated by large eddy simulation (LES). With the correct calibration, those peak values coincide with k_ip;peak ~0:07u²_b. (Figure Presented).