This manuscript aims to experimentally investigate the three-dimensional, time-averaged wake flow of a VAWT farm, both with and without wake control. By utilizing volumetric particle tracking velocimetry, the study clearly identifies how the mean velocity field of the VAWT wake varies with the pitch of the blades. The effectiveness of the vortex modes is highlighted by analyzing the available power coefficient (f_AP), even though the power output of the rotors was not measured. The wake flow topology is thoroughly described and compared for three different blade pitch angles (0° and ±10°). It is important to note that the non-zero pitch values used in this study are relatively large and may not fully replicate in-situ conditions. 

While this analysis is quite interesting, especially for examining the physical aspects of wake flow re-energization, as well as for mean wake modeling, several technical questions are raised throughout the article. 

1. I agree that the wake produced by turbines during energy extraction affects the performance of downstream turbines, making the arrangement of turbines a crucial factor in influencing flow behavior. However, to highlight the impact of both vortex generator modes, the authors have chosen to consider an in-line turbine arrangement. Don’t the authors believe that this arrangement is particularly advantageous for analyzing the wake effects related to the present blade pitch?

Previous analyses (Shen et al. 2024. Ocean Engineering, 311, 118965; Azadani 2023. Ocean Engineering, 272, 113855 and references therein) did not observe that in-line turbine arrangement is the best arrangement for optimized farm performance. Please discuss on the present turbine arrangement choice.

Would vortex generator modes be as efficient for turbine staggered arrangement? Please comment.

2.    The Reynolds number is quite small compared to in-situ conditions. Is there any Reynolds number effect on the turbulent VAWT wake flow? Also, mean flow representations look like laminar-type flow. Please comment.

3.    3D Measurement method and post-processing.

The mean flow analysis relies mainly on the efficiency of the flow measurement method and its associated post-processing method. 

a) Line 257-258: ‘Given an overlap of 75%, the final grid spacing of the velocity vectors is 19mm’ The overlapping mesh grid resolution appears to be quite high compared to the flow scales being considered, particularly near the rotor, which has a chord length of 30 mm. What is the validity of the resulting time-averaged mean velocity field in the near wake of the vertical axis wind turbine (VAWT)?

What about the spatial smoothing effect over the voxel on the mean results?

Globally, what is the size of the smallest resolved turbulent flow scales?

What are the limitations of present velocity measurements?

b) Could the authors detail the computation of the ‘mean’ w_x and w_z vorticity? The resulting vorticity field seems to approach a laminar-type flow field. 

c) The authors present a volume analysis of the energy equation for which the fluxes of the Reynolds stresses Ux’Uz’ and Ux’Uy’ are determined.  Could the authors discuss the determination of the Reynolds tensor components? What is the effect of the spatial flow structure smoothing in the voxel on the fluctuating velocity results? 

d) Could the authors provide an estimation of the error of quantities plotted in Figures 19 and 20?

e) Can the present velocity measurements account for turbulence? If so, a key factor that affects rotor performance and blade fatigue is the turbulent kinetic energy. Could the author display and analyze this quantity in the wake flow of the VAWT as well as in front of the rotors?

These issues need to be addressed to ensure the reliability of the results.  

  4.  Figures 14-15: While the turbine induction process is clearly indicated in front of the first turbine, it seems not to be visible for the other turbine apart from the second turbine with beta=-10°. Is it a consequence of the measurement method and post-processing tools? Please comment.

   5. Mean velocity analysis. 

Numerous analyses of mean flow were conducted to examine the recovery of the wake and the available power. Instead of simply plotting spatial average quantities, it would be more effective to provide additional representations of the mean streamwise velocity over the rotor area, 2D upstream of the rotors. This approach would allow us to assess not only the overall impact of inflow shear on the development of the rotor wake but also its possible influence on blade structural fatigue. Inflow shear is expected to cause asymmetric loading, which could consequently affect the fatigue of the blades. Consequently, alongside the analysis of the mean available power, discussing the predictability of turbine lifespan in relation to inflow shear would be valuable.