|The paper "Observations and explicit modeling of isoprene chemical processing in polluted air masses in rural areas of the Yangtze River Delta region: radical cycling and formation of ozone and formaldehyde” by Kun Zhang et al presents a modelling study of isoprene chemistry at a suburban site near Shangai, China. |
Many of the concerns expressed by the reviewers of the first version of the paper have been addressed, but there remain a lack of clarity as to the objectives of the authors and the limitations of the modelling approach.
The authors use a somewhat limited set of parameters (CO, NOx and 55 VOCs) to calculate the concentrations and the budgets of O3, HCHO and HOx radicals during a field campaign in spring and summer. There are some obvious limitations to such an approach:
1) many important parameters were not measured, including some that the authors themselves indicate as very important: for example, CH4, HONO, photolysis rates, aerosol surface area. There is no attempt to estimate the uncertainty of the model results derived from these missing constraints. In addition, the instrument used to measure NOx has known intereferences, especially when it comes to NO2. This can affect the results but it is not mentioned at all.
2) O3 and HCHO concentrations are affected by several non-chemical processes, which are difficult to account for with a zero-dimensional model. Although dilution, "aloft exchange" and deposition are mentioned at some point, they are not addressed in the discussion. It is legitimate to use such a model to focus on the in-situ photochemical pathways that form and destroy O3 and HCHO, but then it should be made very clear that the analysis is limited to those processes and to local conditions.
3) It is hard to assess the reliability of the model results without radical measurements, and in the absence of a sensitivity study. As such, the radical budgets (Fig. 5 and 6) are qualitative at best. Previous studies may help with an estimate of the level of agreement between model and measurements, but only to the extent that the models being compared are similar. Moreover, there is some discussion of nocturnal processes, but the results of night-time chemistry are even more uncertain than those of day-time chemistry, because the model is not necessarily good at predicting NO3, as other modelling studies supported by in-situ measurements have shown.
4) I don't think the authors have made a compelling case that the chemistry in this area is driven by isoprene, if it is the intention of the paper to demonstrate this. Sure, isoprene is important and affects the formation of ozone and formaldehyde, but does it really dominate over all other VOCs, under these conditions? Before focusing on isoprene, the authors need to show evidence that other VOCs do not contribute as much to the oxidative capacity.
All these uncertainties and limitations of the modelling approach should be clearly acknowledged and openly discussed, and the conclusions should take that into account. Likewise the aims of the study should be clearly stated.