This work consists of one of the first attempts to determine the structure of turbulence in forest areas using fiber-optic distributed temperature sensing against the traditional eddy covariance method. In my opinion, the technique is promising and, in the future, it can help in research involving turbulent exchanges between the forest and the atmosphere, considering that the temperature field is probably what controls these exchanges in these environments. I believe the paper has scientific merits and is within the scope of the journal.

This paper deals with a new technique to measure the turbulence inside canopis using a new technology. The authors described the idea/new instrument, put it at the their lab (outside field) for adjustments and then installed it inside a forest. They did not find good/nice results with the new apparatus and concluded that it has to be improved. Attached there are many others smallcomments/suggestions.

The authors present a very challenging approach to measure wind and sensible heat flux with fiber optic distributed temperature sensing. Unfortunately, the results of the study are not so promising as it appeared that the shroud did not work well to measure vertical wind speed. I appreciate the authors braveness to submit (partly) failed experiments. Also failed experiments can help the community to learn. Having said this, I think the current manuscript needs major revisions as it seems more a good draft then a full paper (yet). Problem statement and method section miss essential information and certain choices are not well explained. Additionally, I think that also many of the figures can be improved with less abbreviations and make them more self-explanatory (e.g., by giving the figure titles as 'setup 1, setup 2'. this would also reduce the caption lenght). In the attached pdf I commented in detail. Here I only indicate my main comments.

1. The outline of the study is that the authors first investigate several shroud configurations on a grass field (EBG). The 'best' shroud is then later used in a follow-up experiment in a forest. However, in the method section there is barely any information on the different setups and why shroud color, mesh size, rigidity or shape would affect the measurements. What were the design criteria. This part should be extended and improved.
2. Why are not all the shroud experiments (EBG) compared to the sonic (thus also setup 1 and 2)? Now the benchmark is the 'unshrouded' FODS measurements, which is also an experimental method. I would benchmarkt the shrouds to the sonic as this is likely closer to the truth.
3. Base the first test, the authors pick 'the best shroud' setup to apply it in a forest. Only surprising change, is that 'suddenly' the shroud lenght is increased. While from study 1 the authors could have learnd that dimensions matter for the wind direction. This is in my view a major shortcomming of this paper.
4. Despite the admitted 'failure' of the forest experiments, the authors still show the initial plan to calculate the sensible heat flux. But what is the value of this, once the wind speed measurements are not correct?
5. The reference list contains 34 references, from which 16 are from the own research group. This is almost 50%! I highly recommend to put the study into a more broad context. Many other groups also worked in this study field, including groups that also work with FODS. In the attachement I added some suggestions.

I hope my comments help to improve the manuscript.