Your manuscript can be published in AMT after you make the following, mostly technical, corrections. Congratulations.

Please consider shortening the introduction. There are repeated statements of what the aim of the study, that VOCs potentially interfere with the NH3 measurements etc.

One of the main conclusions is that "The results of this study clearly demonstrate the advantage of CRDS compared to the performance of the photoacoustic analyzers studied by Liu et al. (2019) for which severe VOC interferences on ammonia were observed." I suggest to phrase this a bit more cautious. Strictly, the advantage would only be proved if the two instruments were measuring NH3 side-by-side in a controlled setting with known H2O, VOCs etc. I think your work convincingly shows that the CRDS suffers from interference of these species to a limited extent, but it is not necessarily saying that PAS would fail if tested in the same Danish stable as you have done here. My suggestion is to nuance the statement. This can be done by pointing out that in your experiment the CRDS had only few interferences, whereas a PAS sensitivity test in comparable (?) circumstances, showed much more interferences from non-NH3 species.

P1, L10: "has been" --> have been
P1, L31. Period missing after Nielsen reference.
P4, L26: Rong et al.
P5, L4: "expect for" --> except for?
P6, L42: (100 pbb) --> 100 ppb?
P7, L14: "errors on few ppb" --> errors of a few ppb would have a small (?) impact

The Associate Editor