Response to Comments by Referee #1

[email protected] & [email protected] We would like to thank the reviewer for careful reading of the manuscript and the constructive comments and questions which have helped us to improve the manuscript. The reviewer comments are given below together with our responses and changes made to the manuscript. General Comments 5 This manuscript describes nighttime observations of OH and HO2* radicals during PRIDE-PRD 2006 and CAREBEIJING 2006 field campaigns and their comparisons to theoretical model simulation results. While HO2* was in fair agreement, observed OH levels were significantly larger than simulation. After detailed discussion on the possibility of artifact, the authors suggested possibility that the combination of additional ROx production and the recycle of OH from ROx through HO2 10 could bring the model simulation in fairly good agreement in OH. For processes, the authors implied that the additional ROx production would be explained by ozonolysis of unmeasured reactive terpenes at ample amount, or more possibly by the downward mixing of PANs as reservoir of radicals. The topic is well suitable to the scope of the journal. Although large uncertainty needs to be taken into account, the analysis is made on the best knowledge basis, with clear logical flow. Two 15 major comments are that (1) the modeled (and observed, if available) PAN concentrations should be mentioned together with the estimated flux, to convince readers, and that (2) discussion on the resemblance of the night-to-night variations in the observed and modeled radical concentrations should be involved. Overall, I recommend publication after revisions with respect to the points raised above and the following minor comments. 20 Response (1) In the revised manuscript, we compare the modeled PAN concentrations together with the estimated flux as suggested. Furthermore, we show measured PAN data for PRD (for Beijing, experimental PAN data were not measured at the field site). In the revised text and revised Fig. 12, we show that the modeled PAN and MPAN concentrations are plausible by comparing with observed 25