Bodini et al. provide a manuscript which describes the observed conditions for a benchmark which was performed in the context of an IEA Wind Task. Since the outcome and interpretation of the benchmark strongly depends on the understanding of the condition during the benchmark time period, this manuscript is of high relevance. The manuscript describes the wind park and wind turbine configuration and puts a focus on the meteorological conditions, respectively the flow features at the experimental site. Although the manuscript is well written in general, I think there are parts which are too vague and require some more explanation. Since the manuscript is clearly put into the context of the modeling benchmark, some more explanation should be given how the described curiosities in the flow features are incorporated in the benchmark and how data is processed and provided within the benchmark. I suggest major revisions in this direction before publication of the manuscript in WES. Below, I list general and specific comments.

General comments:

- The description of the meteorological conditions is very much limited to a local description of the wind. I think it would be important for this Part 1 paper of the benchmark description to include a bigger picture synoptic scale description of the meteorological situation and put it in context with statistics at this site. Is this a "typical" condition, or a rather special one?

- How is the complex dataset and meteorological information finally provided for the benchmark? What were the recommendations for its use? 

- Without having participated in the benchmark, I do not completely understand the context and the relevance of the analyses of the flow features for the benchmark. For example: Was the terrain index used in the data processing and a map with "corrected" elevation values provided? Based on all the measurements, what are the boundary conditions that models used to set up their simulations?  Was there a recommendation or best practice advice? 

p.2, l.45f: instead of RANS "or" actuator disk, i suggest to say "RANS incorporating the actuator disk method" if that is what is meant here.

p.2, l.54: I am not convinced that WRF really bridges the gap between the above-mentioned methods. It actually serves the purpose to connect wake modeling with numerical weather prediction and thus atmospheric features, thus closing an open flank in most other models.

p.5, l.118 and Fig.2: Is there any way to quantify how much these generated thrust and power curves differ from the actual wind turbines in the field? this could be helpful to understand uncertainties, even if it is only for a dedicated operation point which occured during the chosen case study.

p.7f, Tab.2&3: It is great to have such a detailed overview of instrumentation, but how was it prepared for the benchmark participants. They just choose what they find suitable to set up their models? Maybe you could separate it between input data and validation data? 

p.9, l.139ff: The criteria for the selection are mostly availability criteria, with the exception of wind direction. In the next section it is written that there are no "adverse synoptic conditions", but were there no criteria for a minimum wind speed, a wind speed range, or similar during the selection process already? I assume that some variety of atmospheric stability conditions was desired as well, wasn't it?

p.9, l.153: what is a "double LLJ"? Maybe rephrase to "two distinct LLJ features" or similar. Please also give a short explanation what is the forcing mechanism for those LLJ. I assume the first is a nighttime LLJ due to inertial oscillation. The second maybe similarly so, but originating from a different area and being advected over the area of interest after a shift of wind direction? Since you are presenting the meteorological conditions in this Part 1 paper, please include some synoptic and mesoscale description of the situation.

p.9, l.155: From Fig. 3 it seems that only at site A2 and in the weakening phase of the LLJ does the "nose" intersect hub height, but at A2 there are no wind turbines, so why do you claim this could effect power production?

p.9, l.161 and 169: How exactly is TKE derived from the lidar? The peak in TKE is during a transitional phase. Have you checked stationarity during the averaging period for TKE estmation? I wonder if this is rather an instationarity effect than actual TKE. Also, there is a data gap in the vertical profiles exactly during that period. How does that relate?

p.10, l.170ff: There is a lot of speculation about a hydraulic jump and wave breaking. If you include this in the discussion I would expect some more evidence. You may have this looking at the vertical velocity of the lidar measurements, but it is not shown or explained.

p.12, l.197ff: Please give more details on how the EOF was implemented. The database for it is the single, but full day of 23 August?

p.13, l.209: How are the stability classes defined? Simply positive or negative Obhukov length for unstable, resp. stable ABL? Please let the readers know who may not be familiar with the concept.

p.13, l.217f and Fig.8: the terrain index is equal to z_{rel} and presented in units of meters? Please explain and use the units in the figure.

p.16, Fig10: Wouldn't it make sense to make a bias correction for the lidars to remove the discontinuity in the profile?

p.17, l.242: I think the meteorological representativeness remains to be shown. The authors did not give any context of the site statistics.

The comment was uploaded in the form of a supplement: https://wes.copernicus.org/preprints/wes-2026-33/wes-2026-33-RC2-supplement.pdf