This paper carries out different numerical simulations to investigate the influence of different ground/soil structural changes (so called the seismic metamaterials according to this paper) around a wind turbine on the seismic wave-path between the wind turbine and the seismic station. It’s a very interesting and meaningful research in the reviewer’s opinion. The authors investigate four main types of different metamaterial scenarios including cross-shaped holes, half circular trenches at short distances, empty half circular trenches at large distances, and different topographic effects. The research is new, the discussion is insightful, and the English writing is clear and straightforward.

Three minor thoughts the reviewer would like to share:

1. It seems this paper only studies the seismic signal emitted from one wind turbine. Will the investigated structural changes have similar influence on the seismic wave-path due to multiple wind turbines (or a wind farm)? It will probably make the research much more complex, but it may be worth discussing more, especially for the case of empty half circular trenches at large distances. Usually there is a wind farm at a large distance from a seismic station (or seismic stations).

2. If we think in a different way, will the structural changes help mitigate noise if these structural changes are close to the seismic station but not the wind turbine.

3. Is it safe for a wind turbine if digging a 5-20m depth trench which is only 10 m distance from the wind turbine? Well, this may be beyond the scope of this research. It is only the reviewer’s curiosity.

Some technical corrections:

1. It may need to change the “windturbine” to “wind turbine” in the title;

2. Page 3, line 83, “distributed in a within” should be “distributed within”;

3. Page 7, line 157, “of the WT as a point sources” should be “of the WT as a point source”?

4. In the fourth line of the caption of Figure 6, there are two “is”.

5. Page 12, line 249, “This in is contrast” should be “This is in contrast”.

General remarks

The paper is well written and easy to follow, therefore in my opinion there

is no need for major changes in the text.

The authors theoretically evaluate possibilities to reduce the influence of

wind turbines (WT) in seismic records by placing empty or filled cavities into

the travel path of seismic noise waves. The efficiency of the the noise

reduction depends on the shape of the cavities and on the observed frequencies.

When focusing on the relevant frequencies between 1 and 10 Hz which are

observed in many previously published examples at distances above 1 km from

the WT it turns out that the most effective noise suppression (among the

investigated models) is achieved with empty half-circular trenches of

certain dimensions. Regarding the ongoing discussions on the necessity of

restrictions of WT operations in the vicinity of seismic stations this is

a valuable contribution to the search for remedies in this conflict of interests

between the operators of seismic stations and the operators of WT. It shows

that there exists at least a theoretical possibility of noise suppression,

however, the practical feasibility remains unclear and this probably cannot

easily be answered by seismologists. Below I collected some

questions concerning this topic. Details need to be further investigated

also by engineers and constructors of WT.

Some part of the paper deals with the influence of topography on seismic

wave propagation. The result of this section, however, does not lead to

unique recommendations concerning the possible locations of WT if the geologic

structure of the whole area is not really well known. So there might be

noise reductions or enhancements depending on the velocity relations

between the mound-like structures and the ground below. Also the 

frequencies affected depend on the dimensions and shape of the

hill structures and are not easily predicable. Only in specific cases the

location of a WT on top of a hill does help to reduce emitted noise. It is good

to know that topography may influence the propagation of noise but the

mentioned uncertainties and the fact that in many cases the location

of WT is restricted by a number of site selection conditions will most

probably not make topography a decisive factor in noise mitigation.

To give an example, one could have a look at the station GR.GRB1

(https://opentopomap.org/#marker=13/49.3903/11.6506) where 5 WT are

installed in an approximate distance of 3 km to the north-east of the station

(marked in the opentopo map). The WT are located on top of a small mountain

range with an elevation difference of about 100m to the station level. The

noise spectrum of GR.GRB1 shows one of the largest WT noise peaks at 1.1 Hz

observable at German permanent stations:

https://www.szgrf.bgr.de/cgi-bin/send_windspec.py?station1=GRB1&station2=None&year1=2017&year2=None&night1=Night&night2=Night&minfrq=0.9&maxfrq=8.0&lopsd=10.0&hipsd=1.0e5&linlog=linlog&smooth=3&operation=overlay&shownumbers=nonumbers&submit=Create+Figure

Of course, this does not prove anything, but it is an example that even with

WT installed on top of a hill there may exist very large noise signals.

Specific comments/questions

What is the radius of the half-circle trenches? This is nowhere explicitely

mentioned. The reader just can guess from the "distance of the WT from the

trench" that it is 10m.

As far as I know the concrete foundation of a WT can have dimensions (diameter)

in the order of 20m. That means that the trench is right at the border of

the foundation (if not within). I guess this cannot be realized as it affects

the stability of the WT and therefore I see problems to install

circular trenches in proposed dimensions. What radius would be necessary

in practice instead? The trench should be significantly decoupled from the

foundation otherwise there will be no mitigation effect.

I understand that using a point source makes things easier and when looking

from large distances principal effects may be well described by such a

simplification. However, the foundation of a WT plays a major role in

the generation of seismic waves by its movement as a whole in different

directions. If the source dimension and the dimension of the wave blocking

cavity around it are in the same order, how much can we rely on the results

using a point source approximation?

Can you descriptively quantify the mitigation effect somehow? For example,

how many more WT with trenches could be installed in comparison to one

without to have the same effect at the station? Or how closer a WT with trench

could be situated to a station compared to one without?

Since water in the trenches seems to foil the mitigation effect, the trenches

need to be properly protected against intrusion of water (from below and above).

On the other hand the two walls of the trench should be decoupled from each

other as much as possible to prevent vibrating of the trench stabilizing

construction as a whole. This seems to be a quite challenging task for the

constructors of such an installation.

I could imagine that readers would be interested in the computing costs

of your simulations. It is of course strongly dependend on the hardware,

but giving an idea on the computing times would be helpful.

Technical corrections (giving line numbers):

25: affected range 1-10Hz, correct for distances above 1km or so. At smaller

    distances higher frequencies will be observed.

40: tress -> trees

45: larges -> larger

48 *have* shown

114: comma missing

116: "placed 10m in front of the trench": is this the center of the (half-)circle?

152: "as in the first scenario" -> "as in the scenario with cross shaped holes"

196: reduces -> reduce

204: efficient*ly*

figure 3a) and 3b)  diferent -> different (2 times)