General comments

This paper provides an extensive demonstration of the use of vortex identification methods applied to particle image velocimetry measurements. In addition, various  numerical differentiation schemes for those schemes using velocity gradients are tested and analysed. Vortex location, vortex core radii and jittering areas are discussed extensively.

The paper is well organized and is clear to read. The introduction reviews literature articles concerned with vortex identification and describes the aims and the paper structure. One of the major criticisms about this paper is the novelty. The methods utilized are very well documented in the literature and adopted in various studies. Their limitations and advantages, including the numerical schemes are quite well known by now, considering also sister literature concerned with helicopter rotor aerodynamics. The authors need to expand more on the novelty of the article. Refer to the specific comments.

Section 2 details the VIM approaches used in this paper and the accompanying numerical schemes. This is well written.

The experimental data used in this work is then described in section 3. The authors do not provide any comprehensive detail on the uncertainty of the experimental measurements which could have an important impact on the uncertainty of the results. Uncertainties of numerical schemes are well discussed but this is not enough to ensure a lucid discussion on the observed divergences between schemes. A discussion on the uncertainties derived from the PIV measurements should at least be made.

Further comments on the results section are expanded upon in the specific comments but in general the results are well presented. The only criticism here is that the conclusions provided are still not fully convincing in light of the above mentioned shortcomings.

In conclusion, the paper currently lacks novelty and requires a more robust consideration of experimental uncertainties and how these could possibly impact the presented results. Nevertheless it is felt that this work presents a worthy analysis which, if refined further, can provide an incremental step in the application of VIMs.

Specific comments

While the paper provides very interesting insights on the use of the different VIMs for vortex tracking, the novelty of the article is still unclear. The advantages of the Graftieaux’s approach, which is hailed by the authors as the most suitable approach are already known. The authors should expand more on the novelty of their article in the introduction section since it is for now not very apparent.

The effects of experimental uncertainty are lacking in the paper. The SPIV data has some level of uncertainty in the results and how this gets propagated in the methods that are being proposed is not very clear. At the very least, a discussion of these effects should be made. This could also have an impact on the Graftieaux approach used. The literature on the uncertainty on PIV should be thoroughly reviewed in order to be able to support your discussion of the effects.

Figure 10 – It is recommended to remove the line plot and use a scatter plot for such representations.

Line 264-267 – The following sentence should be clarified further “Alternative methods such as the prediction from time series vortex locations might be also successful using the rest of the schemes due to the small discrepancy between the vortex center locations between VIMs and schemes; however, more than one vortex age is needed.”

Figure 17 – For completeness the figure needs a colorbar.

Line 333 – It is not very clear whether the authors are rejecting the uneven shedding effects on the observed double peak results.

Line 343 – Do you here mean for vortex kinematics analysis or do you really mean that the methods are simply not suitable for establishing both position and motion of the tip vortices? “In fact, both schemes ignore information either forward or backward from the grid on the implementation of differentiation. Therefore, they are not suitable for vortex analysis.”

Conclusion - For the most part it is felt that the issue of the double peak has remained unresolved in this work. The authors seem to attribute these to purely numerical artefacts. Do the authors feel confident about this conclusion? Could experimental uncertainties also be responsible for this?

Grammar

Line 25 – “It is shown, by using the vorticity to identify the vortices, a high variation in the position of the tip vortices.”

Line 237 – Change “are originated” to “might originate”

Line 269 – Sentence structure is poor here: “the Graftieaux 24-points as well only vorticity magnitude cases are presented.”

Line 365 – “The two peaks found in the jittering…” – please rephrase

The authors presented an interesting study in particular in presenting vortex identification methods and their application in PIV data to analyze tip vortex structures of wind turbines. The results are quite informative in terms of identification of vortex core, vortex location and vortex jittering. The manuscript is well designed in structure and clearly presented. To this reviewer, the manuscript is worth being published. However, there are some serious issues need to be further clarified or revised.

General comments:

• The authors should explain their contribution to the field more clearly. The paper definitely needs more description in terms of its novelty and how it distinguishes itself from previous literatures. In particular, the methods and their applications have been already addressed by other researchers; hence, the authors should demonstrate their contribution.
• In the introduction section, the authors addressed different PIV measurements performed by previous investigators particularly those focused on tip vortex flow. However, to this reviewer, there are more studies, also worked on the behavior of tip vortices, that can be included in the literature review.
• The authors have presented an extensive description of VIM methods which predict vortex behaviors. It would be very informative to include analytical approaches such Rankine model which predict tangential velocity profiles of vortex and compare your results with those that can be obtained from those models.
• The authors are well familiar with the fact that there are two governing parameters, i.e. local Reynolds number and tip speed ratio that affect the flow structure of the turbine including tip vortices. The authors need to discuss further about the role of tip speed ratio in their assessments of vortex location, vortex core radii and vortex jittering.
• The authors have employed the results obtained from a PIV measurement to perform their analysis. However, they should present more specifications of the PIV test such the sampling rate of the measurement, the phase phase-lock process and the number image pairs per second for each azimuth angle of the blade.
• The authors should provide enough information if during the measurements the turbine was subjected to any blockage effect in the wind tunnel or not. Regarding that, they should calculate blockage ratio of the turbine based on the tunnel cross section area and considering the tip speed ratio at which the turbine is performing, they should discuss whether the turbine is experiencing blockage effect. If the blockage effect is high, it would affect the experimental results including the velocity field and wake expansion (which also determines the vortex location) significantly.
• The authors should demonstrate more clearly that how the convection velocity has been estimated, particularly from the PIV data. Did you consider the sampling rate of the measurements for each azimuth angle of the blade? How did you make sure that you are tracking the same vortex as moving from one image pair to the next one?

Specific Comments:

• Page 3, line 70: what is Θ_M?
• Page 7, line 145: what is the tip speed ratio of the turbine? Is it smaller or bigger than the design tip speed ratio?
• Page 7, line 150: More clarification about the experiment set-up and process is required, such as the sampling rate, frequency of the laser and camera as well as error analysis.
• Page 7, line 155: How was the process of phase-lock measurements performed?
• Page 7, Figure 2: the location of the camera is not clear in the figure.
• Page 10, line 195: what is v(x,y)? it is mentioned that v(x,y) is induced velocity; however, at line 185 induced velocity is represented by u’(x,y). Which one is the correct one? It is confusing.
• Page 11, line 230: It is mentioned that “the presence of the multiple maxima and the ring-like distribution of the parameters ω and Q can be explained through different hypothesis. On one side, the cause could be the level of noise in the vortex core because the lack of seeding.” If this can be one of the reasons, why you do not get the similar behavior in Figure 6?