This is a very thorough study of a promising approach for establishing a stochastic reference frame for the clumped isotopes of nitrous oxide.  As the paper states, better understanding of the nitrous oxide atmospheric budget is extremely important due to its contribution to global radiative forcing.  In general, quantifying the atmospheric budget is an under-determined problem and the use of additional isotopic species is likely to provide useful constraints.   This will be most useful if highly accurate measurement techniques are developed. This, in turn, requires effective calibration methods and a stochastic reference frame is critical to achieving this for clumped isotopes.  Hence, the topic is appropriate to this journal.  Furthermore, the approach is creative and builds effectively upon previous work. And the results are substantial. The authors demonstrate a very promising approach for creating a stochastic reference frame for two of the three clumped species of interest and for SP. Even the null result that the 15N15N clumped species does not equilibrate is an important and useful finding.  This paper should certainly be published in AMT with a few minor modifications.

My only significant recommendation is that the authors should do a better job of assessing their progress toward the ultimate goal of having a sufficiently accurate reference frame.  That is to say, is equilibrium achieved with sufficient accuracy to match the measurement uncertainty?  It seems, at least in some cases, that this is not the case (See Fig 4 and Table 1) since at line 180, the authors state that the measurement uncertainty “ranged from 0.09 ‰ to 0.18 ‰ for all isotopologues”.  If complete equilibration is not achieved, then the authors should say why not and how significant is any remaining systematic variation and what they propose to do about it in future work.  If, on the other hand, they believe that they have achieved complete randomization within the experimental error, they should state that more clearly and quantitatively.

Finally, I have a few additional comments and questions.

• At line 165, unspecified decomposition products were separated from N2O. How was this accomplished and what were the likely products?
• At line 88, the authors begin a description of their fast dual-inlet system for trapped gases and cite two references. They should also cite Wang et al. (Anal. Chem. 2020, 92, 2, 2034–2042), which first described this innovation.
• The temperature dependence of the catalyst activation shown in Figure 3 is extremely useful.
• The paper is well written, and the authors do give proper credit to related work. The abstract is clear and complete, and the language throughout the paper is fluent and precise. The paper is appropriately referenced.