This is a well written paper which addressing a topic of interest which has been little explored. I would suggest that it can be published with some minor changes:

• Line 229: not sure what is meant by the 'bi-modal velocity deficit distribution'. Please elaborate.
• Figure 12b: it would useful to include the measured power output for rows 2-3 also. I understand that there may be commercial confidentiality issues, but presumably if the values are normalised as for row 1 there should not be a problem?
• Figure 13: make it clear that the values are simulated (I presume)?
• Line 512: it is not obvious that there are patches with little or no turbulence in the TKE1.5 scheme which are much different to the other schemes. Maybe this could be highlighted on the plots?
• Line 566: edit the superfluous text from the acknowledgements

There are a few typos:

• Figure 2 caption line 2: should be 'outlined', line 3 should be 'is outlined'
• Line 146: the Obukhov lengths should have units of metres
• Line 180: should be 'number of particle'
• Line 266: should be 'parameterization'
• Line 374: should be 'maximum' (I think?)
• Line 453: should be 'microphysics'
• Line 477: should be 'turbines are less...

The manuscript describes a case study of interactions between internal gravity waves (a bore initiated by nearby convection) and wind turbines in a wind farm, in Oklahoma. The case study uses a two-domain nested simulation, with the horizontal resolution down to 20m in the nested domain. Although the simulations present several difficulties (sensitivity of the convection and ensuing bore to the microphysics scheme, fine-scale processes, eg wakes, to include in the wind farm) the authors show convincing simulations, supported by the careful comparison to the observations. Publication after minor revisions is recommended.

Minor comments:

l34: there is a case study by Ralph, Neiman and Keller, 1999, worth mentionning with respect to fronts generating gravity waves near the surface

l39-49: repetition: '... in the US Great Plains. In the US Great Plains...'

l63: affects -> affect

l124: it would be useful to indicate the local time already at ths stage of the manuscript; it is done later on of course, but the reader wonders when first encountering the time...

l146: the three numbers are odd, for the Obukhov length: the unit is not given, and two numbers are negative...

l156-157: what is meant by 'or in the operational mode..'?

l198-199: the formulation is a bit surprising: 'the setup is unique (...) only two domains.' This is presented rather as a strength of the study, but using only two domains implies a strong ratio between the resolutions used in one and the other domain, which is generally considered unfavorable...

l215: the formuation is ambiguous: "0 hour forecast (which includes the hour..." If data has been assimilated up to that point, it is odd to call this a forecast.. this shoudl be reformulated

figure 4: it is somewhat disturbing to use the same colormap for the two panels, but with different ranges

figure 12: plural: 'power output[s]'

l455: repetition of 'results'; vary the choice of words: simulations? flow variables?

l463: 'For the 20 m domain... '  formulation can be improved...

figure 14: the power output in the circules (white to red colors) is only indicated for the three central turbines... why?

l476: 'by up to 56 %': it is good to give the extreme value, but other statistics (mean) could also be meaningful to include

Ralph, F. M., Neiman, P. J., & Keller, T. L. (1999). Deep-tropospheric gravity waves created by leeside cold fronts. Journal of the atmospheric sciences, 56(17), 2986-3009.

This paper reviews the effect of a thunderstorm-downdraft generated atmospheric bore and associated gravity wave on a downstream wind farm in central Oklahoma. The wind farm was located in the American Wake Experiment (AWAKEN) field campaign domain. Overall, the paper is well written, the methods cogently presented, and conclusions sufficiently supported by the analysis. It would be interesting to see what effects, if any, the wind farms had on the downstream propagation of the bore/gravity wave, but perhaps the topic for another paper. A brief discussion of the "forecastability" (short-term) of such events, particularly in the context of power production, would have been illuminating (although other PECAN/AWAKEN papers are addressing this?). Specific comments:

Figure 1: may help to have a larger map (perhaps on the scale of the state of Oklahoma) to provide some geographic perspective. I understand the need to capture the location of individual wind farms/turbines, but this is shown in Figure 4.

Table 3, and lines 365 et seq.: although it is demonstrated the gravity waves eliminated the LLJ, did the jet subsequently recover given there were several more hours until sunrise, and the "After" window only covers the period immediately after the waves have passed? This would also relate to the general vertical thermodynamic structure, and, of course, power production at the wind farm.