Interactive comment on “Development and application of a reactive plume-in-grid model: evaluation over Greater Paris” by I. Korsakissok and V. Mallet

and model’s scores The RMSE improvement presented in the abstract corresponds to the RMSE on stations, given in Section 4. We agree that the model’s performance cannot be summarized by this mild improvement, and we should insist more on the better representation of the spatial variability. As the reviewer states, the results for NO in terms of total RMSE are “surprisingly low” when given in the abstract without the general context. As explained in the paper, this is due to the low impact of point sources in total NOx emissions (about 20%, since most emissions are due to traffic). In this regard, using a plume-in-grid model for point sources cannot have a very high impact on the global statistics, for this case-study. We removed the RMSE scores from the abstract. Using NOx We also addressed the reviewer’s interesting suggestion, to use NOx instead of NO. The RMSE results are similar (the part of point sources in total emissions is still low), and are therefore not shown. We also used NOx in the sensitivity study, and we thought interesting to complete the results with a figure which summarizes the sensitivity to each parameters (diffusion, time step between puffs and injection time) of SO2, NOx and O3 (Section 5.1.2, new Fig.13, and attached figures). It represents the RMSE difference (Polair3D plume-in-grid, μg m−3) for the base case, and for three sensitivity simulations. It emphasizes the parameter to which each species is the most sensitive (green bar on the figures). Comment on Section 1 See general comments. Previous applications of such coupled models on photochemistry focused on how to model chemical interactions. In this study, we chose to apply the up-to-date chemical model of SCICHEM model, which gives good results, and to focus on other aspects, mainly the influence of diffusion and local-scale modeling. The application also differs from previous uses of plume-in-grid models, as we explain in the introduction.