On top of the three reviews that were provided by the anonymous referees there are a number of issues I, as handling editor of the manuscript, would like the authors to comply with before final publication can take place.
My remarks and requests do not have directly to do with the scientific content of the manuscript but more on the way the “message” is presented.

As already addressed with the first version of the manuscript a decent structure was absent. And that has not improved a lot trough the revisions. This means I would like you to restructure the manuscript before it can be published. Below you will find my view on how to provide the manuscript with a (much better) structure, which I am afraid, you have to comply with.

The manuscript contains (the use of) on site measurements, analysis of measured data, numerical simulations as well as validation (comparison of measured data with numerical predictions).
When intensive use is made of measured data in a scientific manuscript, as is the case here, it is a good habit to include a materials and methods section somewhere in the first part of the paper. This section does not necessarily has to be “branded” or titled “materials and methods” but its goal is to provide the reader with a good understanding of the experimental set-up, prior to diving into the results.
The present version of the manuscript presents measured data (flap-wise moments) together with a comparison to simulated data (in fact simulated data from two different approaches) already in the introduction section. And this is not an ideal approach to say the least.

The best way to deal with it is to start with an improved yet much shorter introductory section. This section should contain the problem definition (large scatter in the determination of loads when using turbulent inflow conditions from met masts) as well as a proposed approach to mitigate the scatter. Then introduce possible way(s) out for this problem and then suggest to use “local” inflow conditions taken from local sensors instead of met mast measurements. Here you may introduce the DanAero project at a glance as a good and available source of data to investigate this possible way out.

Subsequently a (literature study) section with a discussion of advantages and disadvantages, as well as the differences that one can anticipate on already upfront should be discussed. This may be largely based upon existing literature.
E.g. issues like the importance of taking into account induction velocities near the rotor (plane) leading to local reduction of axial velocities but also the rotational sampling effects, a.o. the rotational sampling of turbulence.
According to the present article the impression is given that this is “discovered” during the research and then subsequently discussed on page 13. But in fact this phenomenon is already known since decades in the wind energy research community (under the title “rotational sampling”). It is the effect of the change in turbulence spectrum seen when moving with the rotation of the blades: the spectrum of inertial turbulence is rotationally sampled by the blades, generates spectral peaks at the rotational speed and at higher harmonics, and hence is inherently present when blade mounted devices capture local turbulence.
In other words this rotational sampling of turbulence should be discussed in the beginning of the paper. Not only as a scientific feature as such (in the literature section above), but later in its practical implementation.
Questions like: how this rotational sampling is implemented in reconstructing the inertial inflow. If, and if so how this is implemented in (the pre-processor of) HAWC2, because this may affect the comparison of the use of the met mast to the use of blade mounted sensors.

Then the “materials and method” section:
Current sections 2.1; 2.2 and 2.3 should be combined with the parts of the current intro section that deal with the met mast description and measurements. Also issues regarding validity of the use of met mast data (according to the standards suppose) should be presented, as well as the five hole pitot tube, calibration procedures and further measurement issues.
The second part of this section may then deal with information extracted from the measured data, such as the inflow characteristics, the turbulence intensities, turbine tower loads etc. So this is again a combination of information at present presented in the Introduction section and the current Method section (to a large extent this is found in the present section 2.6).

Next section in de new manuscript then describes the cases with possible deviations of the general process of data acquisition and processing as described above.
Then comes the description of the simulation model. In view of the rotational sampling issue mentioned above also something must be said about how this is dealt with in the present simulations. Both in the met mast case as in the pitot tube case.
Then the comparison of both approaches gets the floor followed by a conclusions section.

Other issues to be dealt with in order to get it published:
It is needed that more quantitative information is provided regarding accuracies. Phrases like “relatively accurate” and “some discrepancy” should be supported with at least an uncertainty quantification (which may be based on references, such as e.g. from Pedersen et al).

Furthermore the information presented in the current version of the manuscript regarding lidars for BMFS can be discarded, since it does not provide any additional information relevant for the purpose of this article.


I hope the authors will comply with this direction, so that a positive decision for publication can be taken.

Your paper is now accepted. However, please don't use "met mast" in the abstract but "meteorological mast" since you cannot find the former in a English dictionary. In the main text you should define it the first time you use it.
Thank you for publishing with WES, Jakob Mann