Dear Matthias Karl and colleagues,
Thank you for taking on board the referees' comments and the other contribution, and for making such a thorough re-working of your paper. I am happy to accept it now for ACP, on consideration of the technical issues listed below,
Sincerely,
Rob MacKenzie
Abstract: I don’t think the referees and other commentators are asking you to back away completely from your study… I would suggest “Under inefficient dispersion conditions, the model predicts that condensational growth contributes to the evolution…”
Abstract: Perhaps the following would avoid giving a false impression of accuracy of the parameterisation: “The simplified parameterisation of aerosol processes predicts the change in particle number concentrations between roadside and urban background within 10% of that predicted by the fully size-resolved MAFOR model.”
Page 2, para 2. Most readers will know perfectly well what you mean but you might consider changing the description of aerosol process models to make sure that the adjective Lagrangian applies to the fluid dynamics not particle growth/evaporation dynamics. I’d suggest “Aerosol dynamic models (i.e., process models of aerosol microphysics, often employing Lagrangian approaches to the fluid flow) have been used…”
Page 2, section 2.1, para 3. Is there a fossil here? I thought you did now consider fractal geometry in your sensitivity studies? Also mention of neglect of Van der Waals and viscous forces in the next para? Perhaps start this para with “In the default model run, …”
Section 2.2. It is best to state which one (spatial?) dimension is included in MAFOR (downwind distance?)
Section 2.2. If I understand correctly, all the observed data is fed into the model as either roadside or urban background size distributions. Given the dominance of dilution in the model, it is forced by the dilution scheme to ‘relax’ towards the background size distribution with a time constant dictated by the dilution scheme parameters. I think it is important to state this in section 2.2, and perhaps again in the results section, so that the unwary reader is not tempted to take Figure 4 as a validation or evaluation of the model. To evaluate the model would require some data to be held back for that purpose – again, perhaps it’s best to say that this has not been carried out in this paper, which focuses on the model’s responses to changes in its treatment of aerosol microphysics.
Section 2.4, immediately after numbered list. It’s a complicated sentence but I think there is a noun-verb agreement issue: shouldn’t it be “…data… were obtained simultaneously…”? Strictly, data is always plural, so the next sentence should also say “were obtained”.
Table 3. Perhaps use * etc to designate table footnotes, to avoid tripping up unwary readers of the 6th column?
Bottom of page 5. It would be helpful to report the mean bias between the SMPS instruments as well as the r-squared correlation.
Section 3.2. Here you say “the aerosol dynamics box model”, meaning MAFOR presumably. Is there an inconsistency here with the earlier statement about spatial dimensionality? See also the Conclusions, where the model is one-dimensional again.
Section 3.2 (or before). Have I missed the definition of the edges of the largest and smallest size sections in the model? If this is dealt with in the SI, please make reference to that in the main text. I’m thinking of this here because I am wondering whether the results in Figure 4 show the whole model size-domain and, if so, what the boundary conditions are, especially at the small particle end.
Immediately below eqn 4. We can say simply “Refer to section S3 of the Supplement for details of the implementation.”
Conclusions. I would avoid use of the word ‘evaluated’, because of the issue of held-back observations discussed above. How about saying you have investigated?
Conclusions. Instead of “We coupled…”, I would suggest “We used a simplified treatment of the dilution of an air parcel from roadside to urban background to relax the model towards observed background size distributions.”
Final paragraph could be polished as follows:
“Designing mitigation policies for ultrafine particle pollution in the future will require operational modelling of PN on urban scales. The simplified parameterization we present can be implemented in both Gaussian and Eulerian models. However, it is recommended that such modelling systems are evaluated against measured PN and correlative data in a variety of urban settings.”