Dear Authors,

I'm glad to inform you that your manuscript has been considered suitable for publication after some minor revisions, as detailed by the referees' reports (see below).

Kind regards

Gianfranco Vulpiani

-----------------------------------------------------
Report #1
The authors have thoroughly addressed concerns and remarks from the reviewers. As a consequence, the paper has been modified and improved considerably.
However, I believe that some other corrections and/or clarifications are needed before publication.
The specific comments are as follows:
1. The exact period and dates of the comparison between the two AWS should be emphasized in the text. It is only reported in the abstract, while, in line 102, it is worth wondering whether these dates correspond only to the scale of the scatterplots or are instead the actual period of comparison.
2. Clarifications appear necessary about the general evaluation of the 3D-Printed WS. Also, the shortcoming of the durability of the low-cost sensors, as well as of the 3D-Printed WS, should be highlighted.
I mean: the comparison lasted eight months. Temperature sensor began to fail in month seven; RH sensor in month six; Atmospheric pressure sensor was replaced at the beginning and then suffered some communication problems. Wind speed sensor worked well until month 7, whereas wind vane was problematic for the entire period. Solar radiation sensor suffered from different bias (before high and then low). Rain gauge tended towards high bias starting from month seven.
Because of this, some conclusions appear to be pretty daring. Some of the low-cost sensors worked very well but for a limited period of the comparison. Furthermore, 3D printed parts gave some concerns in terms of durability. Hence, the limited time in which these low-cost sensors can serve as viable alternatives to commercial weather stations should be pointed out.
3. Line 193: You mean the maximum rain rate of the heavy precipitation event, I guess. If so, it should be added.

Typos:
Line 47 “are”
Line 79: “Raspberry” for consistency
Line 127: “are”
Line 161: “extend”
Line 216: “is”
Line 229: “low-cost” for consistency
Line 233: “is”

-----------------------------------------------------

Report #2
The Authors presented a revised version of the paper, which incorporates a large part of the suggestions supplied in the first round of the revision process.

They performed a good work, although the analysis section still remain confined to a basic level. For example, no additional focus on a specific relevant meteorological event has been performed. The authors cited only a heavy rainfall event (in which a rain rate over 100 mm/h has been detected by both sensors), but they did not present a detailed investigation producing new figures and/or new discussions.

Considering the novelty of the topic, I suggest to accept the paper after a minor revision. The latter involved mainly the Figure 10, which is difficult to read. In my opinion, the authors should produce a wind rose plot which may be a good solution to synthetize the difference in performance between the 3D-Station and the reference Mesonet sensor.

Moreover, in the caption of Figures 5-13, the authors must clarify the difference between the dashed blue line and the solid black line. I have some doubts about the realiability of the linear regression model of the Figure 9. Please verify the suitability of this linear fit to the data. Moreover, in fig. 9 I see many points placed along y-axis, meaning that calm wind (0.0 m/s) has been measured by 3D-Station; I suppose that those points represent cases of 3D-sensor fault. I suggest to eliminate these cases from the scatter plot and to reproduce, for this figure and also for the other ones, only the cases in which the two sensors worked properly.